Citation
The merging of organizations

Material Information

Title:
The merging of organizations an exploration of the impact on job burnout, stress related illness, and turnover
Creator:
Kent, William Patrick
Publication Date:
Language:
English
Physical Description:
v, 153 leaves : illustrations, forms ; 29 cm

Subjects

Subjects / Keywords:
Consolidation and merger of corporations -- Psychological aspects ( lcsh )
Hospital mergers -- Psychological aspects -- Case studies ( lcsh )
Job stress -- Case studies ( lcsh )
Burn out (Psychology) ( lcsh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references.
General Note:
Submitted in partial fulfillment of the requirements for the degree, Doctor of Philosophy, Graduate School of Public Affairs.
General Note:
School of Public Affairs
Statement of Responsibility:
by William Patrick Kent.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
|Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
25695884 ( OCLC )
ocm25695884
Classification:
LD1190.P86 1991d .K46 ( lcc )

Full Text
THE MERGING OF ORGANIZATIONS:
AN EXPLORATION OF THE IMPACT
ON JOB BURNOUT, STRESS RELATED ILLNESS,
AND TURNOVER
by
William Patrick Kent
B.S., Regis College, 1976
M.S., Washington State University, 1977
,A thesis submitted to the
Faculty of the Graduate School of the
University of Colorado in partial fulfillment
of the requirements for the degree of
Doctor of Philosophy
Graduate School of Public Affairs
1991
>> s
I At,1
$ i


This thesis for the Doctor of Philosophy
degree by
William Patrick Kent
has been approved for the
Graduate School
of Public Affairs
by
Robert
Gage
is
1 \ 12.o/q)
Date
John Buechner


Kent, William Patrick, (Ph.D., Public Administration)
The Merging of Organizations: An Exploration of the
Impact of Job Burnout, Stress Related Illness, and
Turnover.
Thesis directed by Professor E. Sam Overman
The impact of mergers on organizations has been
documented in the literature. One impact of a merger
is stress and the employee/organizational consequences
of that stress. This study examined attitudinal and
behavioral outcomes of the merger process and the
stress related to that process.
Two acute care hospitals that had decided to
merge together were studied. Data from two employee
surveys, turnover data, and employee medical claims
information were analyzed. The results of four
research questions developed for this study were
examined. Analysis of variance and chi-square tests
were used to test the difference in the means of
Maslach Burnout Inventory (MBI) scores and turnover
rates respectively.
The first research question asked if MBI scores
for employees who filed stress-related medical claims
would be different than employees who did not file
claims. The results showed no significant difference
between the mean scores of the two groups. The second


research question, which tested for difference in
turnover rates between the hospital that was relocating
in the merger to the site of the other hospital, did
provide data that showed significant difference between
the two hospitals over a four-year period. The third
research question investigated for a possible
association between employee turnover and medical
claims. It was found that employees who filed
stress-related medical claims had a higher rate of
turnover than employees who had not filed claims. The
fourth and final research question analyzed the MBI
subscale scores of both hospitals. The hospital MBI
scores were not significantly different between
hospitals at the time of the first or second survey.
Hospital specific MBI scores were significantly
different:between the first survey and the second.
This study offered partial confirmation for the
merger stress impact models that propose that merger
stress does impact employees and organizations.
The form and content of this abstract are approved. I
recommend its publication.
Signed!
E. Sam 0ve rman


CONTENTS
CHAPTER I
INTRODUCTION...................................... 1
Problem Statement............................. 4
Significance of This Study.................... 5
Stress and Illness......................... 7
Mergers and the Impact on Employees..... 8
Definition of Terms........................... 9
Stress...................................... 9
Burnout.................................... 12
Merger and Acquisition..................... 13
Limitations of the Study..................... 15
General Methodoligical Limitations......... 18
Organization of the Remaining Chapters.... 19
CHAPTER II
LITERATURE REVIEW................................ 22
Job Stress................................... 22
Stress and Illness......................... 23
Multidimentional Aspects of Stress..... 27
Organizational Issues and Stress........... 29
Organizational Liability and
Costs Due to Stress...................... 33
Methodoligical Weaknesses
of Stress Research
37


Burnout..................................... 40
Definitions and Descriptions.............. 40
Factors in the Development
of Burnout.............................. 43
The Emotional and Physical
Impact of Burnout....................... 44
Relationship of Burnout to Stress......... 46
The Maslach Burnout Inventory (MBI)..... 49
The Phase Model of Burnout................ 50
The Link Between the Merger Process and
Stress-Burnout, Illness, and
Organizational Issues .................... 51
Impact of Merger Activity on Employees.. 56
Merger Stress Impact Model................ 58
Relocation in a Merger
Process.............................. 61
Summary...................................... 62
Purpose of the Research Questions......... 64
Research Questions......................... 66
Research Question Number 1.............. 66
Research Question Number 2.............. 66
Research Question Number 3.............. 66
Research Question Number 4.............. 66
CHAPTER III
RESEARCH METHODOLOGY OVERVIEW.................... 67
Sample...................................... 69
Characteristics of the Sample............. 72
Maslach Burnout Inventory................. 75
vii


Turnover, Demographic and Other
Information Collected.................... 78
Analyses................................... 79
Analysis of the Four Research
Questions................................ 80
CHAPTER IV
FINDINGS......................................... 84
Research Question One......................... 85
Results................................... 86
Total Burnout Scores........................ 92
Total Burnout Scores for All Groups......... 92
Research Question Two......................... 95
Results.................................. 96
Research Question Three...................... 99
Results.................................... 100
Research Question Four....................... 102
Results.................................... 102
Summary of Findings......................... 107
CHAPTER V
CONCLUSION..................................... 111
Discussion of Results...................... 111
Implications of Findings................... 116
APPENDIX A....................................... 125
APPENDIX B......................................... 129
BIBLIOGRAPHY....................................... 132
viii


TABLES
Table
3- 1 Sample Demographics..................... 73
4- 1 Hospital 1 (hospital relocating)
Group A (employees not filing
claims) and Group B (employees
filing a claim) Compared at Time 1
( 1987) and Time 2 ( 1989)
First Data Collection................. 87
4-2 Hospital 1 (hospital relocating)
Group A (employees not filing claims)
Time 1 (1987) Compared to Group A -
Time 2 (1989) and Group B (employees
filing claims) Time 1 (1987)
Compared to Group B Time 2 (1989)
First Data Collection................. 88
4-3 T-tests for the Difference Between
Employees Who Are Not Filing Claims
(Group C) and Employees Who Did
File Claims (Group D) from Hospital 1
and Hospital 2
Second Data Collection................ 90
4-4 Hospital 1 Matched Groups A and B
(subscales totaled)
First Data Collection................. 94
4-5 Hospital 1 and Hospital 2 Employee
Groups. Group C (employees who did not
file claims) and Group D (employees
who did file claims)
Second Data Collection.................. 94
4-6 Turnover Rate of Hospital 1
(employees relocating) and
Hospital 2 (employees not
relocating )...................... 97
4-7 Turnover Rates of Employees Who
Filed Claims Compared to the
Turnover Rates of Hospital 1 and
Hospital 2
Second Data Collection................. 100


4-8 Hospital 1 (employees relocating) and
Hospital 2 (employees not relocating)
Burnout Scores for Each Hospital at
Time 1 and Time 2.....................
4-9 Hospital 1 (employees relocating)
Compared to Hospital 2 (employees
not relocating)
Burnout Scores at Time 1 (1987)
and Time 2 ( 1989)....................
103
106
x


FIGURES
Figure
2.1 Merger Impact Model ..................... 52
2.2 Merger Stress Process ................... 54
2.3 Merger>Stress>Impact Model .............. 59


ACKNOWLEDGEMENTS
This dissertation benefited from the input of
many individuals. I want to particularly express my
gratitude to Al Johnson and Dr. Kit Tennis. They
provided the technical support and direction that
allowed me to develop and finish this study.
Most importantly, there are four people who
deserve very special thanks for providing encourage-
ment, support and understanding along the way. I would
like to thank Max Suiter, my friend and partner in this
process, who spent many hours refining this
dissertation. Her patience and skill helped to make
this dissertation a reality. My father Al, who always
stressed the importance of education and perseverance,
provided me unconditional support. Finally Barbara, my
wife, and Allyson, my daughter, who allowed me the
I
time, gave me their love, and understood my selfishness
and mood!swings throughout this long journey. To all
of these wonderful people, I give my thanks and will be
forever appreciative.
v


CHAPTER I
INTRODUCTION
"Stress: The Test Americans Are Failing."
In the April 18, 1988, issue of Business Week
magazine, the author made the point that stress is
epidemic in U.S. business. Competition, lay-offs,
mergers and acquisitions are adding to the stress
levels of employees across the nation. The article
quoted mental health experts' estimates that 15% of
executives and managers suffer from depression or
critical levels of stress that will eventually
affect job performance. According to this article,
the cost to business for stress related problems and
mental illness is estimated to be $150 billion
annually in health insurance and disability claims,
lost productivity, and other expenses.
The outcomes of job stress have important
organizational cost implications (Ganster et al.,
1982). Organizational productivity, profitability
and utilization of health care services are all
negatively impacted by job stress (Parker and
DeCotiis, 1983). Ivancevich and Matteson (1980)
estimated that the total cost of stress related


problems, i.e., lack of productivity, absenteeism,
health care costs, etc., may be approximately 10% of
the gross national product in the United States.
Parker and Decotiis (1983) found that job stress
negatively affects organizational commitment, job
satisfaction, avoidance behavior, and job
performance. According to a study by the California
Worker's, Compensation Institute, worker's
compensation claims for stress related disabilities
in California rose from 1,178 in 1979 to 9,368 in
1988 (Peterson, 1990).
As worker's compensation claims have
increased in the 1980's for stress related claims,
so has the merger and acquisition activity of
American businesses. W. T. Grimm & Co. reported
that in 1990 there were 2,074 potential mergers and
acquisitions announced with a value of $108 billion
for the deals in which a price was specified. That
compares with 2,366 announced transactions and a
total announced value of $221 billion in 1989. As
corporations have striven to become more "lean and
mean," they have cut costs and restructured to the
point that employees are having a difficult time
with the. stress caused by the restructuring.
According to J. M. Rosow, President of the Work in
2


America Institute, a not-for-profit research
organization, since 1982 about 4.5 million Americans
at all levels of the career ladder have lost their
jobs because of restructuring.
A frequent result of mergers and restruc-
turings is downsizing. In the August 4, 1986,
Business Week magazine, the author stated that
cutbacks among fifteen of the Fortune 500 companies
accounted for a total of 128,000 lost jobs. During
the 1980s, in the financial industries, 3,400 banks
and savings and loan associations were involved in
mergers affecting the lives of 822,000 people (Dull,
1986).
In the hospital industry, which this study
focuses on, there are predictions that merger
activities in hospitals will downsize the industry
by 17.6% over the next ten years, spanning from 1988
through 1998. As a result of a decline in inpatient
utilization, there will be 480,000 fewer hospital
full-time equivalents. Approximately 20.7% of all
hospitals in the northeast have merged and 29.8% of
all hospitals in the southwest have merged
(Cherskov, 1987; Fink, 1988).
A merger creates a number of potential and
powerful antecedents of stress and stressors.
3


Stressors may manifest themselves in physiological
and emotional symptoms and illnesses (Bruckman &
Peters, 1987; Ivancevich & MattesOn, 1980). Mergers
may contribute to employee burnout. Signs and
symptoms of employee burnout include: bickering, a
sense of resignation, stagnation, a lack of vision,
incompetency, a low level of mutual trust,
frustration, lack of follow-up and support,
inefficient use of resources, poor communication,
fear and unwillingness to take reasonable risks
(Fink, 1988).
Problem Statement
Most of the research regarding the impact of
mergers and acquisition activity are case studies
from the perspective of the managers who have lived
through the experience. The insights and inform-
ation from their first-hand accounts are important
in that they provide suggestions as to a management
course of action during a merger process. There is
little empirical information other than financial
statistics and productivity outcomes related to
merger activity. There is however, a growing body
of literature suggesting that mergers create stress
that has deleterious effects on health, job
4


satisfaction, and work performance (Caplan and
others, 1975; House, 1980; McLean, 1979). Caplan,
Cobb, French, Van Harrison, and Pinneau (1975)
stated that "stress" refers to "any characteristics
of the job environment which posses a threat to the
individual" (p. 3). Others have identified the
effects of stress as problems related to the
physical and mental well being of the worker
(Jayaratne and Chess, 1984; Kahn and Quinn, 1970b;
LaRucco, House, and French, 1980).
Significance of This Study
It is important to develop an understanding
of the financial and human resource impacts of
merger and acquisition activity to an organization.
The loss of productivity, decline in employee morale
and the stress associated with merger activity are
important issues in today's business environment.
The significance of this study will be realized if
four issues can be linked into one model of
understanding: merger activity, stress/burnout,
illness, and turnover.
The study was conducted within a framework
which asserts that there is stress caused from
merger and restructuring activity. This stress, if
not dealt with, will develop in this
5


stress continuum to burnout and manifest itself in
physical and psychological health care costs. Burn-
out is experienced acutely in the human resource
professions, and of particular importance for this
study, in health care facilities (Adams, 1983;
Blanks, 1985; Costa et. al., 1983; Dames, 1983;
Fong, 1985; Grutchfield, 1982; Horner, 1985;
Townley, 1985).
In a study conducted by Boland (1970), fifty
Chief Executive Officers of organizations in the
process of contemplating a merger were asked to rate
the importance of twenty-six factors when investi-
gating a company for possible acquisition. Prior to
a merger, only three of the top twelve factors were
related to the human resource function. After a
merger, the same CEO's rated human resource concerns
in seven of the top twelve factors.
In other studies reviewed by Robino and
DeMeuse (1985), managers of companies acquiring
companies, and the managers of the companies being
acquired, both found the process to be stressful.
Using Caplan et al.'s (1975) definition of job
stress "any characteristics in the job environment
which posses a threat to the individual" and strain
as "any deviation from normal responses in the
6


person" (i.e., a psychological strain such as job
dissatisfaction, anxiety, and low self esteem), it
is logical to seek a possible connection between
merger activity, stress and health care utilization
costs.
In researching the literature for this study
it was not possible to find many studies or research
linking all four issues into one construct. There-
fore, the subjects were analyzed in parts and then
constructed into a model of understanding.
Stress and Illness i
Several studies have shown that perceived
i
job stressors are related to one or more mental
health problems. These problems include
neuroticism, tension, depression, irritation and
anxiety (Beehr, 1976; Beehr, Walsh, and Taber, 1976;
Caplan et al., 1975; Gemmill and Heisler, 1972;
House and Harkins, 1975; House and Rizzo, 1972;
Ivancevich, 1974).
In addition, other studies point to a
relationship between stress and physical illness.
The following studies reviewed possible associations
between stress and physiological responses.
Researchers studied increases in uric acid (Cobb and
Kasl, 1972; Shirom, et al., 1973), blood sugar
7


(Schar, Reeder, and Dirken, 1973), an incidence of
peptic ulcers (Cobb and Kasl, 1972) as they related
to stressful events. Self-report measures of
physical illness related to work included those by
Caplan et al., (1975), House and Harkins (1975),
Jacobson, (1972), Margolis, Kroes, and Quinn (1974),
Mettlin and Woelfel (1974), Patkai, Frankenhaeuser,
Rissler, and Bjorkvall (1967), and Powell (1973).
Other studies have shown the relationship
between stress and utilization of health care
services1. Rahe, Gunderson, Pugh, Rubin, and Arthur
(1972) in a naval study showed that sailors who had
jobs that were judged to be more stressful tended to
report to sick call and require more medical
treatment than those in jobs judged to be less
stressful.
Mergers and the Impact on Employees
Schweiger and Denisi (1991), in their
longitudinal field experiment focusing on
communication issues and mergers, make the point
that companies engage in merger-acquisition
activities because it allows them strategically to
accomplish certain goals. The organizations may
like the compatibility of product lines, access to
8


however, may be extremely costly and undermine the
reasons for implementing the corporate strategy.
Finally, as stated earlier, the merger and
acquisition impact on employees has been reviewed
from first-hand experience and review of case
studies (Bruckman and Peters, 1987; Gilkey and
McCann,1988; Jamron, 1983; Jemison and Sitkin, 1986;
Korman, Rossenblum and Walsch, 1978; Kitching, 1967;
Perry, 1986; Pritchett, 1985; Schweiger and
Ivancevich, 1985).
Definition of Terms
In the management sciences it is common to
use terms that are open to a variety of operational
definitions. It is important for this study to have
clear definitions for the terms that are central to
this document. The terms that need to be defined
and linked together to form a relationship are
stress, burnout, and mergers.
Stress
Selye, who pioneered the concept of general
adaptation syndrome (1952), used the term stress to
refer to a general stress response of the person and
defined stimuli that caused stress as stressors.
The term stressor is used to refer to the environ-
9


ment or stimulus, the term distress to refer to the
person's perceived negative reaction to the environ-
mental stresses, and the term stress response (or
strain) to refer to the physiological and behavioral
consequences of stress. Parker and DeCotiis (1983)
stated that they struggled with the term because it
has been both broadly and narrowly defined. It has
been treated as a stimulus, a response, an environ-
mental characteristic, an individual attribute, and
an interaction between an individual and his or her
environment (Beehr and Newman, 1978; Katz and Kahn,
1978; Levi, 1981). Some researchers labeled
physiological dysfunctions as stress (Ivancevich and
Matteson,11980), and others have called it a
consequence of stress (Schuler, 1980). Ivancevich
and Matteson (1980) have labeled stress as "the most
imprecise term in the scientific dictionary" (p. 5).
Perspectives of stress have included individual
differences, environmental factors and a mixture of
the two. The choice of one perspective over the
other hasitypically been determined by the research
questions to be answered. For example, medical
researchers who are interested in the physiological
outcomes Of stress, treat the individual as the unit
of analysis and focus on the personal character-
10


ristics, such as heredity, age, and personality
traits. Researchers interested in organizational
perspectives focus on job related characteristics.
An integrated approach focuses on both individual
differences and environmental factors (Ivancevich
and Matteson, 1980). The Parker and DeCotiis (1983)
concept of stress is limited to an emotional
response to stimuli that may have dysfunctional
psychological or physiological consequences. By
using the term stress too widely to denote pressure
on the individual (e.g., workload), its effects
(e.g., poor work performance), and also the
individual's reactions (e.g., alcoholic drinking),
researchers have contributed to conceptual and
definitional confusion of the issue.
A result of the lack of definitional clarity
regarding stress phenomenon, many have adapted the
guidelines of Schwartz (1980) who used the term
stress to refer to environmental stimuli, the term
strain to refer to the physiological and behavioral
consequences of stress, and the term stress
management to refer to the changing of any aspect of
the environment or person that will decrease strain
and promote health. Due to their integration of job
stress and merger activity, this study will use the
11


Schweiger and Ivancevich (1985) description. They
state:
Stress is an imbalance between the requirements
to make an adaptive response to some change,
event or condition and the repertoire of the
individual. Stress involves transactions that
trigger many individualized actions, thoughts
and feelings. (p. 48)
In addition to the description by Schweiger
and Ivancevich, Parker and Decotiis (1983) defined
stress as follows:
Stress is the emotional response to stimuli that
may have dysfunctional physiological and
psychological consequences. (p. 165)
Burnout
The definition of burnout used by this study
was first developed by Kahn (1978):
A syndrome of inappropriate attitudes towards
self, often associated with uncomfortable
physical and emotional problems ranging from
exhaustion and insomnia to migraine and ulcers,
(p. 61)
Kahn's description/definition of the burnout
syndrome works well with this study because it
connects the syndrome with physical and emotional
illnesses that are mentioned in the stress
literature.
Maslach and Jackson (1981) developed the
term "burnout syndrome." The syndrome is described
as increased feelings of emotional exhaustion. As
12


their emotional resources are depleted, workers feel
they are no longer able to give of themselves at a
psychological level. Secondly, individuals experi-
encing burnout syndrome also develop a negative,
cynical attitude about the clients with whom they
work. The third aspect of the burnout syndrome is
the tendency to evaluate oneself negatively when one
evaluates one's work with clients. The workers are
unhappy with themselves and dissatisfied with their
accomplishments on the job. The Maslach and Jackson
definition has been accepted widely and appears to
be the definition most commonly referred to in the
literature. According to Maslach and Jackson the
three subscales of burnout, (1) emotional
exhaustion, (2) depersonalization, (3) low personal
accomplishment, are interrelated. It is thought
that depersonalization and low personal
accomplishment are impacted by emotional exhaustion.
It is also possible for a person to experience one
of the three parts and not the other two.
Golembiewski, Munzenrider, and Carter (1983) state
that when all three elements are experienced, the
overall effects of burnout increase in severity.
Merger and Acquisition
Merger is defined as "the union of two or
13


more commercial interests or corporations" and
acquisition is defined as "the act of acquiring,
something acquired, e.g., as an addition to an
established group" (American Heritage Dictionary,
1990). In this study both of those definitions will
serve as an overview of what is being discussed.
The merger process, however, is the description of
what happens to the organization and the employees
before, during and after a merger. The
interpersonal and organizational dynamics generated
by the interaction of events across process stages
becomes the major source of stress and burnout for
employees (Gilkey and McCann, 1988). Mergers can
cause a variety of basic organizational behavioral
problems both prior to and after the transition.
These behavior problems can be a cause of role
conflict, ambiguity, anxiety, antagonism, anger and
fear (Pritchett, 1985). Other symptoms and
behaviors include personal stress, depression,
distrust, absenteeism, turnover, job
dissatisfaction, organizational ineffectivenss,
over-extended management resources and an overall
resistance to change (Pappanastos, Hillman, and
Cole, 1987).
Schweiger and Denisi (1991) assert that
14


uncertainty and anxiety develop within the work
force due to the lack of communication pertaining to
the merger. This uncertainty and anxiety can lead
to dysfunctional outcomes such as stress, job
dissatisfaction and low commitment to the
organization. This in turn may lead to diminished
productivity and an increase in turnover and
absenteeism. It is the merger process that will be
reviewed and linked with the definitions of stress
and burnout.
Limitations of the Study
Data were collected from surveys given in
1987 and 1989. Approximately 1,110 employees of two
urban hospitals were questioned. Data were also
collected; from medical claims information provided
by the hospitals' medical utilization review company.
The data collected from medical claims included data
from January, 1988 June, 1990.
A variety of issues may limit the general
reliability of this study's findings: 1) the
respondents surveyed came from the entire range of
jobs in the institutions: administrators to line
staff, patient care employees to support system
personnel. The type of work provided by some of the
15


respondents fit the burnout literature profile
(e.g., nurses, therapists, social workers, etc.)
while others have little or nothing to do directly
with patients (i.e., plant maintenance and
environmental services); 2) the study also utilized
self-report measures; 3) the self-selection results
may not be indicative of the organization as a
whole; and 4) an average of 53% of the total
hospitals' employees completed the surveys in 1987
and 1989, the remaining 47% may be more or less
representative of the burnout experienced in the
organization.
Another limitation is the use of only one
i
medical claim's data base. The respondents have
three different insurance benefit plans from which
to choose. This study was able to access
information from only one of the plans offered. The
plans have different deductibles and co-payments for
mental health benefits. It is not possible to
discern why any respondent chose a particular
benefit plan. The plans also have different access
systems and provide the delivery of health care with
different models. The percentage of employees using
the benefit plan in the organizations utilized for
this study was sixty-two percent. The other two
16


plans had twenty-two and sixteen percent of the
employees respectively opting for those systems.
The majority of the employees chose the insurance
used for this study because it is a hospital system
plan that provides a good price and more options for
care. The other plans are either more expensive or
have restrictions as to what method of health care
delivery one can receive. Due to the studied
hospitals' human resource department's access to the
hospital insurance plan system that the majority of
employees utilize, it was possible to gather data
from that insurance organization. This was not
possible with the other benefit plans due to those
companies not being owned by the hospital
organization.
The study's inability to access data from
each of the insurance plans offered by the hospital
system does raise the issue of validity regarding
the samples of data used. It is possible that the
thirty-eight percent of the employees who did not
choose the insurance plan utilized in this study
could have a higher incidence of stress-related
claims and therefore different results in the data.
The benefit structures of the plans are different
and the delivery of care is provided with distinct
17


health care mechanisms. The distinction between the
plans may or may not induce certain individuals, who
either had pre-existing stress-related illnesses or
were unaware of a problem, to choose one plan over
another. These issues do point to questions
regarding the validity of the sample. The approach
of this study is to rely on the fact that the vast
majority of the employees (62%) did choose the
insurance plan used in this study.
The last limitation of this study is that it
utilized a single method research design. Although
there are many advantages to multiple data col-
lection methods, this study did as many as it was
able to do because of the expenses in terms of time
and resources. Resource and organizational
limitations made further data collection prohib-
itive .
It is important to note that "time horizon"
may play a significant role in research concerning
mergers and acquisitions. Buono and Bowditch
(1989) suggested that as much as two years may be
needed before the full impact of the merger activity
can be assessed.
General Methodological Limitations
The use of correlational design does not
18


answer the question of causation and one cannot
discern whether the medical claims are a result of
stress from the job or from another variable. It
also does not allow one to make the statement that
merger activity itself or other factors in the
external environment are the reasons for the burnout
scores. Although data were collected at Time 1
(1987) and Time 2 (1989) it was not possible to
indicate the direction of the causality between
burnout, medical claims and turnover rates because
of other issues that may play a role in the dynamics
of the three issues over time (e.g., nursing
shortages, leadership changes, combination of events
in a changing industry).
The limitations of a correlational design
are, however, partially offset by the use of
longitudinal data and the opportunity to utilize
extensive data from a large organization consisting
of two separate hospitals. The sample of the data
and the information used is not normally available
outside of applied research settings.
Organization of the Remaining Chapters
Chapter II contains a study of selected
literature related to stress, burnout, and impacts
19


on employees and organizations experiencing the
merger process. This chapter is divided into three
major topics: (1) review of job stress and how that
stress is related to illness and health care
utilization, (2) burnout syndrome, and (3) the
relationship between merger activity, stress and
organizational behavioral outcomes. The chapter is
. I
designed to provide the reader with an overview of
linkages that make up the thrust of the study and
also present the research questions of interest.
Chapter III focuses on the method of study
utilized in the research. This chapter is divided
into five major sections: (1) provide an overview of
the research objectives and method, (2) discuss the
characteristics of the sample used in this study,
(3) discuss the processes of the data collection,
(4) discuss the measures used in this study, and (5)
discuss the methods of analysis utilized in
examining the data.
Chapters IV and V, respectively, represent
the findings and the conclusions of this study.
Chapter IV addresses each of the research questions
examined in the study and presents the results of
the statistical techniques used to analyze the data.
A brief summary and discussion of the results for
20


each research question are provided. Chapter V
concludes with a discussion of the implication of
these findings. It also offers suggestions for
future research related to the dynamics of merger
activity as it relates to burnout and stress related
health care utilization, and organizational
behavioral outcomes.
21


CHAPTER II
LITERATURE REVIEW
Job Stress
In reviewing the literature regarding
stress, one could start with Selye's (1952) work.
He saw a wide variety of environmental events or
"stressors" as producing a single specific pattern
of bodily reaction. House (1974) stated that the
pattern begins with an initial lowering of bodily
resistance during which a variety of infectionous
diseases develop, but under normal circumstances
would be successfully resisted. The process acti-
vates various bodily defense mechanisms character-
ized by arousal of the autonomic nervous system:
adrenalin discharge, an increased digestive reaction
coupled with lower levels of resistance to
inflammation in the stomach (and other bodily
tissues). If this bodily state is prolonged, it
results in a wide range of what Selye determined
"diseases of adaptation" (e.g., cardiovascular and
renal diseases, rheumatisms and arthritis, ulcers,
inflammatory and allergic diseases, etc.). This


review of Selye's work says simply that some
diseases are caused by the body's own attempts to
adapt to stress rather than by any external agent
indirectly.
Stress and Illness
Holmes and Rahe (1967) have shown that
changes in the person's life situation from whatever
cause (marriage, occupational and geographic
mobility, deaths of significant others, etc.) are
associated with increased incidence of a wide
variety of diseases, both physical and mental. Cobb
(1972) found that air traffic controllers had an
excess of every psychosomatic disease that could be
studied compared to controlled groups of individuals
in less stressful occupations. Cobb compared
thousands of air traffic controllers with a similar
number of pilots. All of them had to meet the same
physical requirements for licensing. When the
incidence of stress related disease of these two
groups was compared, hypertension was four times as
prevalent, diabetes controlled by diet 2.3 times as
high, and peptic ulcer more than twice as high among
the air traffic controllers as among the pilots.
Also, the busier the airport became for the traffic
controllers, the higher the rates of these diseases.
23


In reviewing studies regarding stress and
coronary disease, one finds much support for the
relationship between stress and heart disease.
Russek (1965) studied 100 young coronary patients
and 100 controls. When comparing the two groups
Russek found that "prolonged emotional strain
associated with job responsibility" preceded the
heart attacks of 91% of the patients, while such
strain was evident in only 20% of the control group
Factors such as family history of heart disease,
diet, obesity, smoking and physical exercise were
taken into account. According to the Russek study,
there is evidence that job pressures, such as work
overload, responsibility, and role conflicts,
significantly increased heart disease risk and
actual coronary heart disease.
Dorian and Taylor (1984) studied the
connection between stress factors and the
development of coronary artery disease. They refer
to Jenkins (1971) and his four clusters of
psychosocial factors. The following factors have
been consistently related to coronary heart disease
(1) Type A behavior pattern, (2) sustained, dis-
turbing emotions (including anxiety and depression)
(3) "overload," especially excessive work overload,
24


and (4) socioeconomic disadvantage. Dorian and
Taylor added (5) the risk traditionally related to
lifestyle, (6) acute, stressful or emotional
situations, and (7) mitigating circumstances that
increase or decrease the likelihood of chronic heart
disease development. The authors use the example of
the individuals working in the Kennedy Space Center
during the mid 1960's. The stress due to the race
to put a man on the moon had an impact on the
employee's health. The Space Center employees
worked at a fast pace with frustrations such as
budgetary restrictions and the knowledge that they
would lose their jobs at the end of the project.
High rates of anxiety, neuroticism, and depression
were associated with divorce in three-fourths of the
marriages, the highest per capita consumption of
alcohol in the nation, and a sudden death rate that
was 50% higher than that of age and sex match
control groups.
Weiman (1977), in his review of occupational
stressors and incidence of disease/risk, states:
There is absolutely no doubt that the threats to
the well being of man result in a variety of
physiologic body responses. Fear clearly
elicits tachycardia, dilation of pupils,
peripheral vascular baso construction....
(p. 121)
Occupational adversity can also result in
25


physiological responses which may become irrevers-
ible if the environment never changes or if the
adaptation does not occur. Weiman supported Selye's
perceptions that disease/risks occur more frequently
when workers are either under-stimulated or over-
stimulated .
Tracey (1965), after extensively reviewing
the literature on stress and somatic complaints,
concluded that there is overwhelming evidence and
proof that stress causes, or is linked to disease.
Norfolk (1977) reported that stress has a
significant impact on sleep patterns, psychosomatic
ailments such as tension, headaches, and
indigestion.
There have been many studies regarding job
stressors and mental health. The symptoms in these
studies included neuroticism, tension, depression,
irritation, and anxiety. The following studies
focused on a relationship between job stressors and
poor mental health (Beehr, 1976; Beehr, Walsh, and
Taber, 1976; Caplan et al., 1975; Gemmill and
Heisler, 1972; House and Rizzo, 1972; House and
Harkins, 1975; House and Margolis, Kroes, and Quinn,
1974; Patkai et al., 1967; Sales, 1970; Schar,
Reeder, and Dirken, 1973; Shirom, Eden,
Silberwasser, and Kellerman, 1973), low self-esteem
26


(Beehr, 1976; Margolis, Kroes, and Quinn, 1974),
boredom (Caplan et al., 1975) and psychological
fatigue (Beehr, Walsh, and Taber, 1976; Cameron,
1971).
Multidimensional Aspects of Stress
In addition to various conceptions of
stressors and reactions to stressors, there are a
host of intervening and conditioning variables.
These variables include biochemical individuality,
psychological set, cultural factors, conscious and
unconscious mechanisms of defense, social support,
etc. (Beehr and Newman, 1978). Other researchers
(House, 1974; Kahn et al., 1964) felt that personal
characteristics moderate the relationship between
job stress and employee health.
Porter, Lawler, and Hackman (1975) indicate
that employees' needs and values can influence their
perceptions of the task and its environment. The
empirical relationship between objective and
perceived job stressors is not clear. Caplan, Cobb,
and French (1975), and Gemmill and Heisler (1972)
found that people with certain personality
characteristics (e.g., Type A) are more likely to
perceive stressors in the environment than other
people. ,Rosenman, Friedman, and Strauss
27


(1964, 1966) studied men who were typed A or B
personalities. Type A men, ages 34-39 and 50-59,
had 6.5 and 1.9 times respectively the incidence of
coronary heart disease than Type B men. A follow-up
study from four and one-half years later found the
same relationship of behavior pattern (personality
type) in incidences of coronary heart disease. The
issue of individual differences among people has
been studied using psychometric testing as part of
the data collection. Studies using the Minnesota
Multiphasic Personality Inventory (MMPI) (Bakker and
Levenson, 1967; Ostfeld, Lebovitz and Shekelle,
1964) have shown that before their illness, patients
with coronary disease differ from persons who remain
healthy on several MMPI scales, particularly those
in the "neurotic" triad of hypochondriasis,
depression, and hysteria. French (1973) maintains
that stress is often a function of person-environ-
mental fit. This assumes that the person's
perceptions of his or her ability and personality
interact with his or her perception of the situation
to determine the amount of stress in that situation.
Cooper and Marshall (1976) provided more
support for the relationship between stress, the
environment and illness. They assert that studies
28


by Kahn and Quinn (1970b), and Margolis, Kroes, and
Quinn (1974) suggest that there is a growing body of
evidence that occupational stress is a causal factor
in heart disease and mental illness. They define
occupational stress as negative environmental
factors or stressors (e.g., work overload, role
conflict-ambiguity, poor working conditions)
associated with a particular job.
Organizational Issues and Stress
Behavioral scientists (Argyris, 1964;
Cooper, 1973) have suggested that good relationships
between members of a work group are a central factor
in individual and organizational health. French and
Caplan (1973) define poor relations as:
Those which include low trust, low supportive-
ness, low interest in listening to and trying to
deal with problems that confront the organiza-
tional member. (p. 19)
Important studies in this area are by Kahn et al.,
(1964), French and Caplan (1970) and Buck (1972).
The Kahn and French and Caplan study both concluded
that mistrust of persons one worked with was
positively related to high role ambiguity. This led
to inadequate communications between employees.
Coch and French (1948) examined three
degrees of participation in a sewing factory. They
29


found the greater the participation, the higher the
productivity, the greater the job satisfaction, the
lower the turnover. French and Caplan (1970) found
that people who reported greater opportunities for
participation and decision making reported
significantly greater job satisfaction, low job
related feelings of threat, and higher feelings of
self-esteem. Margolis, Kroes, and Quinn, (1974)
found that non-participation at work was the most
consistent and significant predictor or indicator of
strain in job related stress. They found that non-
participation was significantly related to the
following health risk factors: overall poor
physical health, alcoholic drinking, and depressed
mood. Other researchers found a difference in their
findings if social support was included in the
research design. Caplan (1971) found little
relationship between perceived occupational stress
and heart disease risk among NASA professionals
reporting high levels of social support from
coworkers. Those reporting low social support
levels, occupational stress was significantly and
positively correlated with heart disease risk.
Matsumoto (1970) found that Japan's low rate of
heart disease was not due to less occupational
30


stress, but due to the social institutional support
systems available.
Jayaratne and Chess (1984) examined the
relationship between work stress, strain, and
emotional support. Their results indicate a
negative association between support and perceived
stress and strain. They also found no evidence for
the buffering effects of emotional support. Their
study measures strain in terms of "work-related
strain," i.e., job satisfaction, depersonalization,
emotional exhaustion, and "health related strain,"
i.e., anxiety, depression, irritability, somatic
complaints. Parker and DeCotiis (1983) studied the
organizational determinants of job stress. In their
study, job stress was defined as a first level
outcome of the organization and the job. It is a
feeling of discomfort that is separate and distinct
from second level outcomes which were consequences
of job stress. The second level outcome may include
varying levels of satisfaction, organizational
commitment, motivation and performance. Their
analysis of the data they collected supported the
concept that job stress is multi-dimensional.
Another impact of job stress is employee
absenteeism, turnover, and psychological withdrawal
31


(e.g., lower job involvement, less identification
with the organization). Lyons (1971) studied
perceived role ambiguity among registered nurses and
how it was related to voluntary turnover. Role
ambiguity has been at the beginning of the job
stress literature (Kahn et al., 1964). Role
ambiguity is usually defined as the uncertainty
regarding the expectations of an employee in one's
work role. Ineffective communication from the
supervisor is usually an environmental antecedent to
role ambiguity. Margolis, Kroes, and Quinn (1974)
surveyed 1,496 employed persons. They found that
work overload was significantly related to the
number of symptoms that were indicators of stress:
alcoholic drinking, absenteeism from work, low
motivation to work, etc. In that same study they
found a number of significant relationships between
symptoms that are indicators of physical and mental
health with role ambiguity. The stress indicators
related to role ambiguity were depressed mood,
lowered self-esteem, life dissatisfaction, job
dissatisfaction, and intention to leave the job.
The relationships in both of the above examples were
not strong. Role ambiguity and the anxiety
resulting from employees not sure of their position
32


or importance in an organization is heightened in a
merger or acquisition process. The section of this
chapter that focuses on stress and mergers will
explore the issue in more detail.
The literature regarding stress due to job
environment can be summarized by McGrath (1976) and
Caplan et al., (1975). McGrath (1976) said:
There is a potential for stress when an
environmental situation is perceived as
presenting a demand which threatens to exceed
the person's capabilities and resources for
meeting it. There are conditions where one
expects a substantial differential between the
rewards and costs from meeting the demand versus
not meeting it. (p. 1352)
Organizational Liability and Costs Due to Stress
Quick and Quick (1979) reported that stress
is related to turnover and absenteeism. Others
hypothesized that repercussions of job stress are
low job involvement, lack of concern for coworkers,
loss of creativity, and accident proneness (Schuler,
1980). Employee compensation suits which cite job
stress as a source of psychological and
physiological injury are on the rise (Cain, 1985;
Ivancevich, Matteson, and Richards, 1985; Leavitt,
1980; Novit, 1982). Leavitt (1980) cites the
dramatic increase in successful claims for
psychiatric injury under the California worker's
33


compensation system as an example of the role job
stress is placing on the liability concerns of
American, business. Workplace stressors are also
recognized as a legitimate cause of certain cardiac
and psychiatric illness under the Federal Employee's
Compensation Act. In some jurisdictions worker's
compensation benefits now include compensation for
injuries arising as a result of continued stress.
The California Labor Code has a section that defines
such injuries as repetitive mentally or physically
traumatic activities extending over a period of time.
If the combined effect of the activities causes a
disability or need for medical treatment, there is a
case.
The other trend in supporting claims for
conditions based on psychological and social
workplace stressors is the growing epidemiological
evidence of the relationship between workplace
stressors and pathophysiology. Even though
researchers and other social scientists do not find
the studies regarding job stress to be very method-
ologically sound, the courts have found the positive
correlations produced by epidemiologists regarding
types of jobs and various psychophysiological
conditions, ranging from hypertension and peptic
34


ulcers to cardiac dysfunction as being admissible
evidence. If the incidence of psychosomatic
complaints, absenteeism, or other workplace problems
are unusually high, credibility for the power of the
workplace stressors may be enhanced in the eyes of
the court system.
Novit (1982) specifically studied mental
distress issues and claims. The trends regarding
claims being submitted by workers for mental injury
are increasing. The claims for mental injury have
no discernible physical manifestation in the way
they have been traditionally defined. According to
Novit, the Supreme Court of California has stated
that:
The courts of this state have also acknowledged
the right to recover damages for emotional
distress alone, without consequent physical
injuries, in cases involving extreme and
outrageous inter-personal invasions of one's
mental and emotional tranquillity. (p. 53)
Novit explains that there are three areas of law
where there are trends regarding job stress and
mental injury: (1) mental injury resulting from work
stressors and related factors for which the employer
is liable under programs of worker's compensation;
(2) compeinsation beyond "make whole" remedies for
mental anguish resulting from the employer's
violation of certain anti-discrimination statutes;
35


and (3) emotional distress from intentional acts by
the employer designed to humiliate and degrade.
These areas of the law are compensated for by
financial damages secured in suits under the common
law doctrine of torts. The legal implications for
organizations due to the liability of job stress
situations is great. Even with the lack of an
abundance of methodologically sound studies focusing
on job stress and illness, there is still a need for
organizations to recognize the correlations even
though they are not strong. Legal precedents have a
way of moving corporations and small businesses
towards worker enlightenment. If there is a real
monetary cost to a worker-work environment situation
there will be movement towards correction of that
problem situation.
In summary, the outcomes of job stress have
important organizational cost implications (Ganster
et al., 1982). It has been argued that produc-
tivity, profitability (Weiman, 1977), and health
care utilization (Peters, Benson, and Porter, 1977)
are all negatively impacted by job stress. Parker
and DeCotiis (1983) found that job stress negatively
affects organizational commitment, job satisfaction,
and job performance.
36


Methodological Weaknesses of Stress Research
Methodological weaknesses and issues related
to the job stress literature are many. They include
the studies of the perceptual process of occupa-
tional stress. These studies measure both percep-
tions of the situations and psychological outcomes
with self-report instruments (Beehr and Newman,
1978).
House (1974) concluded in his literature
review that:
Traditional sociological and epidemelogical
research on social factors in heart disease is
most often involved in comparing disease rates
across categories of standard demographic
variables. 'Stress is often presumed to be a
mediating variable in such research, but the
nature of the stress seldom is specified
carefully and almost never measured indepen-
dently. (p. 23)
House went on to assert that only when all variables
that mediate and/or condition the relationship
between objective social conditions and health
outcomes are explicitly conceptualized and measured
*
can we have an adequate understanding of the effect
of social stress on heart disease. Other authors,
in reviewing the literature, have also found many
weaknesses and inconsistencies in the studies
available.
Parker and DeCotiis (1983) in their review
37


found many methodological problems. One large
problem is the multicollinearity inherent in a large
number of inner-correlated variables typically
associated with job stress. Another problem in the
study of job stress is the need to measure both
stressors and stress through the unique view of the
individual. The need for self-report measures
common in stress research is a problem. Beehr and
Newman (1978) pointed out that the popularity of
measuring both perceptions of the stress situation
and psychological outcomes with self-report
instruments is a problem and criticism of the
studies. Due to the potential of response bias
inherent in the methodology, Beehr and Newman
advocate the use of self-report perceptions linked
with non self-reported consequences. These
perceptions should be linked with objective measures
of the situation. The perceptions need to be linked
with physical responses or other variables need to
show the relationship between perceptions and other
self-reported data. Beehr and Newman also point out
that there are a variety of disciplines approaching
the problem but not in the same way due to the
difference in the disciplines. Some are only
partially relevant to job stress. They cite the
38


number of medical studies on stress but are not
specifically related to job stress. There are
studies that are currently being produced by
organizational dynamic specialists, but few
adequately cover the medical aspects of stress.
Beehr and Newman promote an inter-disciplinary
approach to the issue.
In the Parker and DeCotiis review (1983) of
the literature they found that despite the interest
in this subject, there were relatively few reports
of empirical investigations of stress in work
organizations. They found that organizational
literature on stress is dominated by concept papers
with few tests of the concepts presented. The
entire idea of a conceptual paradigm for stress
lacks precision in that the term has been both
broadly arid narrowly defined. It has been treated
as a stimulus, a response, an environmental
characteristic, an individual attribute, and an
interaction between an individual and one's
environment (Beehr and Newman, 1978; Katz and Kahn,
1978; Levi, 1981). Beehr and Newman (1978) stated:
More of the elements of the psychological
consequences have been studied than the
behavioral or physiological consequences.
Nearly all of the studies of the elements of
this facet have some methodological flaws, as
mentioned previously. The most common problems
have been (1) the use of self-report data for
39


measuring both 'cause and effect' variables
(especially in studying the psychological
consequences), and (2) the use of correlational
rather than experimental designs (thus
disallowing strong inference regarding direction
of causation). The behavioral human conse-
quences suffer most from sheer lack of study.
Studies that avoid the major problems listed
here are greatly needed. Also needed are
investigations of possible beneficial human
consequences of stress. (p. 689)
Rabkin and Struening (1976) provide a good
short summary as to what most researchers have found
in this area. They support the view that the
correlation between stressful life events and
illness symptoms, though dependable, typically is
low.
Burnout
Definitions and Descriptions
There is not a universally accepted
definition of burnout. The following review of the
literature on this subject will focus on burnout as
it relates to job stress and illness.
Freudenberger (1974, 1975) was the first to
use the term burnout in describing a syndrome in the
late nineteen-sixties. He viewed the syndrome as a
level of a person's ability to perform on a job.
The person who experiences burnout is physically and
emotionally exhausted due to his/her work. Perlman
40


and Hartman (1982) define burnout as encompassing
three components: (1) emotional and/or physical
exhaustion, (2) lowered work productivity, and (3)
overdepersonalization. Maslach and Jackson (1982)
describe the syndrome of burnout as a syndrome of
emotional exhaustion and cynicism that occurs
frequently among individuals who do "people work" of
some kind. Maslach and Jackson observed that
"helping professionals" had a gradual loss of caring
about the people with whom they work. Over time,
they found that they could not keep up the quality
of the care and commitment needed to be successful
with the types of people they were working with.
This ability to provide a sensitive and caring
approach to their clients is the essence of their
job. The emotional exhaustion they were
experiencing is identified as job burnout. Levinson
(1977) asserted that the major defining
characteristic of burnout is that people can't or
won't do again what they have been doing. Levinson
describes burnout as a special phenomenon that
occurs after people have spent a great deal of
effort, intense to the point of exhaustion, often
without possible results. People in these
situations feel angry, helpless, trapped and
41


depleted. They have viewed the experience as more
intense than what is ordinarily referred to as
stress. Carroll and White (1982) emphasized the
importance of integrating both individual and
environmental factors within an ecological framework
to understand burnout. They view burnout as a form
of ecological dysfunction. Belcastro and Gold
(1982) viewed stress as a more general term and
burnout as evidence of work related stress.
Cherniss (1980a) suggests that burnout is the
psychological withdrawal from work and response to
related excess stress or dissatisfaction.
Emener, Luck, and Gohs (1982) formulated the
following composite definition of burnout:
The burnt-out person is one who has lost his or
her concern and enthusiasm for the organization,
co-workers, goals, and purposes of the job and
the customers or consumers of goods or services
of the organization. Burnt-out persons are
frequent complainers who view all ideas and
suggestions in a sour or pessimistic way. They
are resentful, disenchanted, fatigued, bored,
discouraged, confused, edgy, quick to anger, and
frustrated over items of mild importance or
relevance. Burnt-out person acts and feels that
the world is against him or her. The burnt-out
person is one who experiences the negative
things in a disproportionate amount to the
positive feelings he or she feels about the job.
(p. 3)
Even though definitions for burnout are similar,
there is not one definition agreed to in the
literature.
42


Tennis (1986) reviewed the literature of
burnout and was able to synthesize the information
into three points: (1) the concept of burnout should
be limited to work related behaviors; (2) burnout
leads to decrease in performance; and (3) burnout is
a syndrome from which, in its final stages, the
afflicted individual cannot recover without outside
help or changes in the environmental stressors which
contributed to the burnout state.
Factors in the Development of Burnout
Causes of burnout and an agreement regarding
the causes have been varied and confusing in the
literature. They range from tedium and stress
(Pines, Aronson, and Kafry, 1981), career develop-
ment crisis (Cardinell, 1981), and poor economic
conditions (Crase, 1980) to work overload and lack
of perceived success (Weiskopf, 1980). Rosenthall,
et al. (1983) examined the relationship between work
environment attributes in burnout and found that
work pressure was positively correlated with the
emotional exhaustion scale of the Mashlach Burnout
Inventory. Fong (1985) suggests that workload was a
critical ;factor contributing to burnout for nursing
educators. Roelens (1983) reported that nurses'
perceptions of workload was among the factors that
were significant determinants of burnout. Zabel and
43


Zabel (1982) and Townley (1985) found that workload
was not a critical factor in determining burnout.
Grutchfield (1982) found a significant negative
relationship between age and emotional exhaustion in
studying registered nurses. In the same study
Grutchfield reported that there was no significant
relationship between nursing education, current job
level, and professional status in burnout. Dames
(1983) in a study of female nurses found burnout to
be negatively correlated with nurturance and
interaction and positively correlated with anxiety
and aggression. Dames also found a positive
correlation between the amount of time nurses worked
with patients and the degree of burnout.
The Emotional and Physical Impact of Burnout
Burke, Shearer, and Deszca (1984) stated
that individuals seen scoring higher on the two
burnout scores (MBI and Cherniss Negative Attitude
Change) reported more psychosomatic symptoms, more
negative feeling states, less job satisfaction, a
greater negative impact of the job demands on home
and family, greater intentions of turnover, and
actually exhibited tangible signs of poor health
(higher blood pressure, absenteeism resulting from
illness, and currently taking medications). Maslach
44


(1976) also reported that burnout cases have been
correlated with certain indicies of.personal stress;
people undergoing burnout often increased their use
and abuse of alcohol and other drugs as a way of
reducing tension and blocking out strong feelings of
hostility and depression. Maslach and Jackson
(1981) obtained external validation of burnout by
use of observers such as spouses. The external
validation concluded that there were important
relationships between absenteeism and over-
personalization, psychosomatic illness and emotional
exhaustion, and use of tranquilizers and emotional
exhaustion. Belcastro and Gold (1983) had three
major elements in their hypothesis regarding school
personnel burnout: (1) burnout is hypothesized to be
inherent stress of one's job, (2) burnout decreases
the capacity of an educator to perform, and (3)
burnout is thought to produce various somatic
complaints, physical illnesses, and emotional
problems in professionals. In their study they
found that the group experiencing burnout suffered
somatical complaints at either a greater frequency
or with greater intensity than did the group of
teachers who were not burned out. They had an
overall accuracy of prediction of group membership
45


of 90.9%. It appears that these somatic complaints
are indeed significant discriminating variables
between teachers experiencing burnout and teachers
who were not. They concluded that teacher burnout
represents a condition associated with some
potential health risk. The literature also suggests
that personality characteristics, especially
characteristics which lead individuals to choose
human service professions, are common antecedents to
burnout (Pines, Aronson, and Kafry, 1981).
According to Maslach's studies (1976), health care
professionals are well trained in certain healing
and service skills, but are not often well equipped
to handle repeated, intense, emotional interactions
with people. Professionals suffering from burnout
syndrome experience a gradual loss of caring about
the people with whom they work. As time goes by,
they cannot sustain the level of personal care and
commitment called for and the personal encounters
that are the essence of their job. Burnout, in this
context, is the emotional exhaustion resulting from
the stress of interpersonal contact.
Relationship of Burnout to Stress
Burnout is viewed in this study as part of
the stress continuum. That is, burnout is stress
46


that has not resolved or dissipated in any way.
Selye (1975) used the term stress as a non-specific
response to any number of stimuli. Burnout is
conceptualized as a specific outcome of stress held
by an individual over some period of time (Blase,
1982; Cherniss, 1980a; Harrison, 1983; Ivancevich,
1974; Meier, 1983; Schuler, 1982; Sweeney, 1981).
In this study, burnout is perceieved as the final
step of a progression of steps to cope with a
variety of negative stress conditions. If burnout
is the final step in a progression of steps of
trying to cope with a stress, then it could be
assumed that burnout would be an antecedent or could
predispose the individual who is susceptible to job
stress-related medical illnesses to become ill.
Pines (1983) and Pines, Aronson, and Kafry
(1981) have connected their burnout studies to
stress research. In reviewing these studies, the
authors use the term burnout as part of the effect
stress has on an individual.
Farber (1983) viewed burnout as a result of
"unmitigated" stress. Farber also suggested that
burnout corresponds to Selye's (1956) third stage of
response to stress exhaustion.
Gaines and Jermier's study (1983) viewed the
47


emotional exhaustion factor as the strain resulting
from work stress while viewing depersonalization and
devaluation of accomplishments as coping strategies
to deal with the strain of emotional exhaustion.
Because of the lack of clear definitions for stress,
and its close association with the burnout syndrome,
the literature is not clear as to where each term
stands with the other.
The two terms, stress and burnout, are often
used interchangeably, although stress is a more
general term and burnout may be seen as evidence of
stress. The review of the stress literature such as
Beehr and Newman (1978); French, Caplan, and
Harrison (1982); Ivancevich and Matteson (1980);
Quick and Quick (1984); and Schuler (1980), do not
refer to the concept of burnout. The stress
researchers listed above, however, list the burnout
antecedents or effects in their studies. Tennis
(1986) proposes that, since the authors (referred to
above) did not comment on burnout in their research,
this exclusion of burnout may be interpreted that
the authors support the argument for the distinction
between the two concepts. The difference between
the two appears to focus on the idea that burnout is
a respondent or label to responses individuals have
48


to specific stressful stimuli and stress is a more
generic or general term for many workplace or
environmental characteristics. For example, burnout
may be one outcome of stressors in the workplace
(Maslach, 1976).
The Maslach Burnout Inventory (MBI)
"The Maslach Burnout Inventory is designed
to assess the three aspects of the burnout syndrome:
emotional exhaustion, depersonalization, and lack of
personal accomplishment" (Maslach and Jackson, 1986,
p. 2). Maslach and Jackson view burnout as a
continuous variable ranging from low to high degrees
of experienced feelings. They emphasize that it is
not a dichotomous variable, which is either present
or absent. Maslach and Jackson also state that the
scores for each subscale are considered separately
and are not combined into a single total score due
to the limited knowledge of the relationships
between the three aspects of burnout.
Development of the current version of the
MBI occurred over time since 1978. During this time
the MBI has been administered to thousands of
respondents. Due to the widespread use of the MBI,
it is used as a measurement tool in this study.
49


The Phase:Model of Burnout
The phase model of burnout,, according to
Golembiewski and Munzenrider (1988), attributes
progressive prepotencies to the three MBI subscales
and defines phases of burnout in terms of eight
possible high-versus-low combinations of scores on
the three subdomains. Similar models have been
generated and cited in this study (Cherniss, 1980b).
The phase model, according to Golembiewski and
Munzenrider, provides a way to measure burnout in
large aggregates and allows individuals to be seen
in terms of the degree or level of burnout. Maslach
and Jackson (1986) assert that the phase model
proposes an interesting way of combining MBI scores
but state that the validity of this or other types
of phase models depends on more research. This
study used the phase model as one approach to the
data. The phase model is of interest to this study
because as Golembiewski and Munzenrider (1988)
state:
Advanced burnout implies that the individual
experiences a collection of stressors that
cause so much strain that normal coping
skills/attitudes do not suffice. Thus over
extended, the individual is left vulnerable to
various upsets, including those generated
within the body, as well as those caused by
invading microbes who find the body's normal
defenses have become less than adequate to
protect against internal or external attack.
(p. 68)
50


The Link Between the Merger Process and Stress -
Burnout, Illness, and Organizational Issues
Stress and burnout literature have been
reviewed for this study. The last section of this
chapter will focus on the models found in the
literature linking the concept of mergers to job
stress/burnout and organizational impacts and
issues.
Merger stress models were reviewed and used
in this study as the foundation for the research
questions. Two models were chosen to convey the
essence of this research. It was concluded that
these two presented the logic of this study in a
clear, understandable format. The first model was
presented by John C. Bruckman and Scott C. Peters in
the spring of 1987. In their article titled
"Mergers And Acquisitions: The Human Equation" they
explored the impact of mergers on corporate human
resources. Bruckman and Peters developed a model
(Figure 2.1) which proposed that merger activity
leads to merger stressors, which may result in a
variety of consequences. On the individual level,
the impacts are manifested as physiological and
psychological symptoms. The stressors identified in
the model have been mentioned in a variety of papers
51


Figure 2.1: MERGER IMPACT MODEL
Uncertainty
Loss of identity
Job loss or demotion
Job transfers
Compensation/benefits changes
Power, status and prestige changes
Leadership strife
New rules and regulations
New evaluation criteria
Change in reporting relationships
New employees and coworkers


I
PHYSIOLOGICAL
Insomnia
High blood
pressure
Trembling
Muscle pain
Headache
Increased
accident rate
Increased sick
leave
Increased
alcohol/drug
use
Gastrointestinal
difficulties
COMMON IMPACTS
PSYCHOLOGICAL
Depression
Anxiety
Preoccupation
Fear of the
unknown
Loss of self-
confidence
Lifestyle
instability
Marital/family
strife
Irritability
Inability to
make decisions
'------V---------
CORPORATE
& SOCIETAL
Derails primary
corporate focus
Decreased
productivity
Lowered morale
Increased
disloyalty
Increased
turnover
Higher
absenteeism
Increased
stealing
Sabotage
Increased
wrongful
termination
lawsuits
Reprinted with permission from Employment Relations
Today. Spring, 1987, p. 56: "Mergers and Acquisi-
tions: The Human Equation" by John C. Bruckman &
Scott C. Peters. Copyright 1987 by Executive Enter-
prises, Inc., 22 West 21st Street, New York, NY
I

PAUDADTIi
10010-6904. All rights reserved.
52


and articles listed in the literature review of
stress and burnout. Job uncertainty, loss of
security, trauma, increase in rumors, poor
communication, leadership changes, employee benefit
changes, changes in reporting relationships, and
other uncertainties in the workplace increase stress
for the individual. The physiological effects due
to the merger displayed by employees are also listed
The list includes the stress-related illnesses
documented by others quoted in this study. Bruckman
and Peters list the common impacts they observed as
being a result of merger activity. Due to the data
and resources available, this study was not able to
measure all of the impacts listed by the authors.
Important to the purpose of this study is
Figure 2.2 developed by Schweiger and Ivancevich
(1985). In the article titled "Human Resources:
The Forgotten Factor in Mergers and Acquisitions"
the authors proposed that merger stress is like
other stress in that it "involves transactions that
trigger many individualized actions, thoughts, and
feelings" (p. 48). The merger process has stages
that make up the employee's experience of the merger
They include increased arousal by the individual due
33


Figure 2.2: MERGER STRESS PROCESS
Note: Only la sample of possible factors is presented in each box.
Source: David L. Schweiger & John M. Ivancevich, "Human Resources: The
Forgotten Factor in Mergers and Acquisitions"
Personnel Administrator/November 1985, p. 51
54


to the possible consequences of a merger, an
appraisal,of the situation, and a coping strategy
which can.include a dysfunctional response (i.e.,
anxiety, hostility, grief, alcohol and drug abuse).
Schweiger and Ivancevich also provide a list of
possible physiological responses due to the stress
the merger process produces. The responses include
high blood pressure, migraine headaches and
insomnia.
On the psychological side, the impacts are
depression, anxiety, and overall preoccupation with
the issues surrounding the merger activity. The
impacts make it difficult for employees to perform
well. Other psychological outcomes included by the
authors are lower job, career, and life satisfaction.
Additionally, increased absenteeism, turnover, lower
productivity and other behavioral symptoms are also
common. The models of Schweiger and Ivancevich
(1985) and Bruckman and Peters (1987) are supported
by the literature pertaining to the impact mergers
and acquisitions have on human resources (Fink,
1988; Gilkey and McCann, 1988; Jamron, 1983; Jemison
and Sitkin, 1986; Kaye, 1989; Kazemek, 1989; Korman,
Rossenblum, Arthur, and Walsch, 1978; Magum, 1984;
Perry, 1986; Pritchett, 1985; Schoonmaker, 1969;
55


Schrivastava, 1986; Schweiger and Denisi, 1991; and
Work, 1986).
Impact of Merger Activity on Employees
The stress from merger activity manifests
itself in many behaviors and symptoms (Bruckman and
Peters, 1987; Fink, 1988; and Ivancevich and
Schweiger, 1985). These symptoms or behaviors are
the same or similar to the symptoms mentioned in the
burnout and stress literature. The physiological
responses include fatigue, over or under eating, use
of drugs and alcohol, high blood pressure, migraine
headaches, and sleeplessness. The psychological
responses mentioned in the merger literature sound
very much like the descriptions of employees
suffering from the burnout syndrome. The
descriptions by Fink (1988) regarding the
psychological responsiveness of employees in a
merger include, depression, anxiety, chronic
indecisiveness, vague feelings of helplessness, and
malaise.: Loss of control, a sense of resignation,
frustration, poor communication, and a lack of
mutual trust are manifestations of organizational
difficulties in a merger. The difficulties are
similar if not exactly like those identified in the
burnout literature. Schweiger and Denisi (1991)
56


hypothesized in their study regarding communication
and mergers that the announcement of a merger will
result in an increase in uncertainty, stress,
absenteeism and turnover among the employees of the
organization. They also stated that job
satisfaction and commitment to the organization
would decrease. The results of their study
indicated that there was empirical evidence that
when uncertainty increases there is an increase in
stress and a decrease in satisfaction. Price
Pritchett (1985) makes the point that due to a
heightened state of uncertainty, the employees move
to protect themselves. This may show up in morale,
turnover statistics, lost productivity and a less
profitable organization. The coping mechanisms tend
to be highly self-oriented and less functional as
far as the organizational good is concerned.
Pappanastos, Hillman, and Cole (1987) emphasized
that managers who are oblivious to the human
resource side of the merger process will also not be
aware of the impact on performance. The three
common problems that arise in merger situations,
according to Pappanastos, Hillman, and Cole are: (1)
ambiguity; (2) weakening trust level; and (3)
concern about self-preservation. Constant worry
57


about the future usually results in a higher rate of
job dissatisfaction, depression and anxiety,
absenteeism and finally turnover.
Gilkey and McCann (1988) supported the
viewpoint of Pritchett (1985) and Fink (1988) that
stress produced by the merger process is not only
harmful to the employee but can be devastating to
j
the entire organization. "People problems" are
commonly cited (33% of the time) as a major reason
for merger failure (Gilkey and McCann, 1988).
Psychological reactions to the stress caused by the
merger include behaviors mentioned earlier. The
manifestation of these behaviors have grave
organizational consequences. For example, perceived
job insecurity and anxiety can increase job turnover
with a loss of management and technical capacity. A
decline jin morale impacts productivity, commitment,
and loyalty. Increased absenteeism, illness,
substance abuse, and other stress-related symptoms
are costly to the profitability of the organization.
Merger Stress Impact Model
Figure 2.3 was developed for this study to
provide a model that could be used to give an
overview of what was being researched. The model is
a composite of Figures 2.1 and 2.2 and information
58


Figure 2.3: Merger>Stress>Illness Model
In
VO
Merger
Activity
Organizational Issues Employee Reaction to Organizational Issues Organizational Behavioral Outcomes
Turnover increase Absenteeism increase Wrongful termination suits increase Accident rates increase Increased sick leave
Lack of vision -Change in report- ing relationships Role ambiguity increases Leadership issues Anxiety Morale problems Increased disloy- alty Lower job satis- faction
Power, status & prestige changes New co-workers Poor communication Rumors increase Lower trust of leadership Lack of teamwork Inefficient use of resources Unwillingness to take reasonable Increase in medical claims and health care costs Worker's compensa- tion suits for stress increases
Time and workload
pressures risks Psychological
Culture changes Bickering Somatic complies-
Lack of participa- tion in decision- Sense of resigna- tion tions Depression
making process Lack of enthusiasm Anxiety
Career development Supervision changes Institutional Self-preservation Burnout Alcohol abuse Migraines Neurosis Family problems
goals change Physiologioal
Job security Blood pressure
Overall uncer- Insomnia
tainty Gastrointestinal disorders Coronary disease


found in the literature review. The model provides
the direction of the study and the linkages between
merger activity, stress due to the activity and the
impacts on the employee and the organization. Due
to the complexity and multi-dimensional aspects of
stress and burnout, it is important to provide an
overview of how those concepts are connected to
merger activity. This model suggests causation even
though the study's research questions and purpose is
the exploration of possible associations between the
areas focused on in the literature review. As shown
in the graphic, the merger activity is the variable
that begins the chain of events. The organizational
issues list labels organizational issues that
increase in power and in number as the merger
activity progresses. The employee reaction to the
organizational issues follows. The next part of the
model lists the behavioral outcomes or areas in an
organization that can be quantified. These
behavioral outcomes are mentioned in the stress and
burnout literature. Under the organizational
behavioral outcomes is the psychological and
physiological illnesses that have also been
mentioned in the stress and burnout literature and
were used as data in this study.
60


Relocation in a Merger Process
In reviewing the literature regarding the
merger process, specific issues were identified.
One issue is the relocation of employees. Zweig and
Blake (1986) conducted a study regarding the impact
of the stress of relocation on employee health.
They studied whether employees directly involved in
a move would have higher rates of physician
utilization and morbidity than those who were not so
involved, and whether strong social supports would
reduce the effects of the move. Due to the type of
organizations and the relocation involved in this
study, the work of Zweig and Blake was of particular
interest. The movers were compared with non-movers
with respect to the outcome variables: work
productivity, change in health status, physician
visits, and bed disability days. They found that
movers were no more likely than non-movers to report
a decline in work productivity. Moving was not
associated with increased physician visits but those
employees who experienced relocation subsequently
had an increased rate of bed disability and were
more likely to report deterioration in health status
than non-movers. These findings were substantiated
even though the movers were no more likely than non-
61


movers to identify high stress at work. These
findings suggest that the event itself, independent
of its perceived desirability or stress, affected
health. McFarlane, Norman, Streiner, and Roy
(1983), ;Mueller, Edwards, and Yarvis (1977), and
Vinokur and Selzer (1975) suggest that undesirable
events account for much of the health effects of
life changes. Others support the notion that events
do not have to be perceived as undesirable or
negative to pose a detriment to health (Holmes and
Rahe, 19!67; Rahe and Arthur, 1978).
Summary
In summary, the themes that are important to
consider are: (1) mergers cause stress for
'i
employees; (2) this stress is manifested in
psychological and physical medical symptoms; and 3)
there are organizational behavioral outcomes.
Uncertainty and change aspects of the merger process
may be the prime contributors to the increase of
stress to the organization and the individual.
In reviewing the literature regarding job
and occupational stress, stress and the burnout
syndrome, stress and illness, and merger and
acquisition process stress, one is struck by the
62


similarity of the information. The interchangeable
nature of the definitions, the lack of clear
perimeters between terms, and the constant use of
the same examples point out the following: (1) many
authors are attempting to make the same connections
and causal references without sufficient empirical
evidence; (2) the definitions and terms are not
clearly defined and accepted; and (3) the obser-
vations and anecdotal information provided by many
researcheirs and authors is consistent but has little
quantitative support.
The overlapping and similarity of the
research pertaining to stress and burnout is
striking. Beehr and Newman (1978), Emener
(1979) , French, Caplan, and Harrison (1982),
Ivancevich and Matteson (1980), Maslach and Jackson
(1982), Sakharov and Farber (1983), and Schuler
(1980) list organizational structural character-
istics and symptoms (i.e., manifestations) of stress
and burnout that are also cited in the merger and
acquisition literature. The organizational
structural characteristics include lack of
participation in decision making, low job security,
communication problems and loss of control.
Symptoms of stress and burnout health consequences
63


include anxiety, depression, high blood pressure,
insomnia, and increased use of alcohol or drugs.
Organizational behavioral manifestations include
loss of productivity, absenteeism, and turnover.
Purpose of the Research Questions
The job stress, merger stress, and burnout
literature suggest that there is an association
between the three concepts. In reviewing the
literature of each subject, the same terms,
symptoms, and outcomes were mentioned. This
similarity points to an interaction of the three
that overlaps and crosses the boundaries of each.
Further, the organizational structural antecedents
to each of the issues are also similar. The
methodology of the studies and papers written by the
researchers in each of the areas reviewed also point
out many of the same research design flaws.
After reviewing the literature and models
related to this study, the individual and
organizational issues present three areas of
inquiry: (1) Is it possible to verify a
relationship between the merger process, the burnout
syndrome and illness? (2) Is it possible to
associate stress and turnover? (3) Is there a
relationship or association between illness and
organizational problems such as turnover?
64


An overview of how the Merger>Stress>Impact
Model (Figure 2.3) is connected to the research
questions being asked by the study may be
conceptualized as follows:
Merger 1
Process
Attitude
Effect
\
MBI
Behavioral
Outcomes
Claims
Relocation
Turnover
The merger process stress will be measured by the
MBI for attitudinal effects. The behavioral outcomes
will provide the objective measure of the impact
merging of organizations have on employees. For
this study, relocation will serve as one aspect of
the merger process that can be singled out as a
specific stressor.
The literature review provides support for
the research questions being examined. The
methodological weaknesses of previous studies and
the need for empirical support for the assertions
made in the literature have been considered in the
development of the research questions.
65


Research Questions
Research Question Number 1
Will there be a significant relationship
between burnout scores and incidence of stress-
related medical insurance claims?
Research Question Number 2
Will there be a higher turnover rate in
the organization in which the employees will be
relocating to another location than the rate of
turnover in the organization not physically
moving?
Research Question Number 3
Will there be a difference in the turnover
rates between individuals who file a claim for
stress-related medical illness and those who do not?
Research Question Number 4
Will the burnout scores be higher for
employees physically relocating in the merger
process than those employees not physically moving?
66


CHAPTER III
RESEARCH METHODOLOGY OVERVIEW
The research design of this study attempted
to utilize subjective (MBI scores) and objective
(medical claims and turnover) measures of the impact
of a merger process of two hospitals. MBI scores
were collected from two surveys. One survey
occurred in 1987 and the other two years later in
1989. Both hospitals had been acquired by a larger
system in 1985 and were in the beginning process of
merging services and consolidating staff. One
hospital (Hospital 1) had been told that it would be
relocating all staff and services in the future to
the location of the other hospital (Hospital 2).
Stress-related medical claims information
was gathered by the human resource department of the
hospital system. The insurance company benefit plan
most widely used by the employees of both hospitals
transferred the claims data to the human resources
department. This provided confidentiality and
enabled the study to code the sensitive information
to ensure confidentiality throughout the study.
The MBI scores of both surveys were used to


compare the MBI scores of employees of both
hospitals who had stress-related medical claims and
employees who had not filed claims.
Turnover rates for both hospitals were
provided by the human resource department of the
hospital system. Four years of turnover rate
information was examined and averaged to show a
comparison of rates for the two hospitals.
The MBI scores were analyzed using three
approaches: 1) each subscale of the MBI was viewed
separately; 2) subscales were totaled together to
provide one score; and 3) a phase model approach was
also taken to examine the possible levels of burnout
among the employees. The turnover data were
analyzed by developing a percentage of employees who
had left each hospital in different years and
dividing that number by the total number of
employees in each hospital. Those percentages were
then averaged over four years.
Due to the small sample size of employees
who had filed claims within the period of time being
studied, a second data collection was undertaken in
an attempt to increase the size of the sample size
of employees who had filed claims.
68


Sample
The samples used for the research consisted
of approximately 1,110 hospital employees from two
large Rocky Mountain area hospitals. The hospitals
together employed an average (over three years) of
2,232 employees. The hospitals were part of a large
national proprietary hospital system. The two
hospitals have the same administrative staff (i.e.,
CEO, COO and vice presidents). The hospitals have
been planning to merge since 1982 and have been
merging clinical departments and support services
during the data collection period.
The burnout data were collected as part an
on-going organizational development effort. The
medical claims data were collected specifically for
this study. Hospital insurance codes that represent
the diagnosis of an illness were used. They are
commonly referred to as I-CD9 (DRG ICD-9 CM Code
Book Procedure Classification) and CPT (Current
Procedural Terminology) codes which allow the
hospital and medical claims personnel to communicate
the illness for which an individual was being
treated. The illnesses that were identified and
coded using the I-CD9 and CPT reference books are
listed in Appendix A.
69


The respondents to the questionnaire part of
the data collection represent all levels of the
hospitals workforce (i.e., administrative, middle
management, support services and patient care
delivery personnel). The questionnaires were
answered in three-hour data collection sessions in
large groups over a three-day period during the
employees regular work time. The instructions for
each session were identical. The questionnaires
were generally self-explanatory. Confidentiality of
responses was guaranteed to reduce anxiety and
unwillingness to answer the questions honestly.
Employees were told that their responses would
provide the organization with information needed to
improve the functioning of the hospitals and the
quality of their work environment. The question-
naires and data collected for the study were
collected in September of 1987 and November of 1989.
This longitudinal approach to the data collection
provides the opportunity for the investigation of
the possible differences between two distinct
periods of time.
It is important to note that feedback was
given to the employees after the data collection in
1987 and a program to correct the problems high-
70


lighted in the data was started at the hospitals.
The same process was repeated after the second data
collection in November of 1989. The hospitals
developed task forces and committees to respond to
the issues raised by the responses of the employees.
The data collection process for the medical
claim information was tedious due to the need for
confidentiality. The human resource department of
the hospitals coordinated their efforts with the
medical claims utilization services for the hospital
system. This provided the needed confidentiality
for coding to take place. Once data were coded, the
analysis could occur without the risk of
confidentiality being compromised. Medical claims
data was collected for two time periods. Medical
claims data for Hospital 1 (hospital relocating)
employees was collected from January, 1988, through
December, 1989. In a second and separate collection
of medical claims data for Hospital 1 (hospital
relocating) and Hospital 2 (hospital not relocating)
employees, information was gathered for time periods
between May, 1989, and June, 1990. Both sets of
data were used in this study.
It is also important to understand that the
CEO and administrative staff would continue as the
71


leadership of the organization after the merger.
This merger was a friendly merger even though the
medical staffs and employees of each hospital had
their own distinct culture and value system.
Characteristics of the Sample
Demographic information of the samples is
provided in Table 3-1. This demographic data high-
lights characteristics that need to be considered.
First, the large percentage of all of the
groups are female and registered nurses. Dames
(1983) and Grutchfield (1982) have emphasized the
significance of the perceptions and expectations of
nurses and other "helper" types of professions. It
is also significant that out of Group B (employees
filing claims) for Hospital 1 (hospital relocating)
the largest percentage of employees had been with
the organization for fifteen years or more. This
may indicate that the length of service at the
institution and experience of the individuals did
not act as a buffer for stress-related claims.
The issue of possible double counting of
data was explored and found to be inconsequential.
The first data collection of MBI scores from
Hospital 1 (hospital relocating) of employees who
did file claims and did respond to both surveys were
72


TABLE 3-1
Sample Demographics
Hospital 1 (hospital relocating)
1987 Total Organization Demographics
675 Respondents
24.0%
59.0%
23.0%
61.0%
19.6%
17.8%
less than one year with the
organization
full-time/day shift employees
received a three-year diploma
described as white female
were registered nurses
described as professional
Hospital 2 ;(hospital not relocating)
11987 Total Organization Demographics
502 Respondents
23.0% five-ten years with the organization
53.0% full-time/day shift employees
25.3% two years of college
63.0% described as white female
2i7.0% were registered nurses
18.0% described as licensed or technical
Hospital 1 :- employees who did not file a stress-related
claim and responded to the 1987 (Time 1) and 1989 (Time 2)
surveys Group A 183 Respondents
2 9.0%
6:9.0%
59.0%
25.0%
30.0%
27.0%
two-five years with the organization
full-time/day shift employees
described as white females
were registered nurses
completed high school
described as clerical
Hospital 1 employees who did file a stress related claim
and responded to the 1987 (Time 1) and 1989 (Time 2)
surveys Group B 17. Respondents
29.0%
63.0%
75.0%
25.0%
31.0%
14.0%
fifteen years or more with the
organization
full-time/day shift employees
described as white female
completed bachelor's degree
were registered nurses
described as professional
73


TABLE 3-1 Continued.
Hospital 1 employees who filed a stress-related medical
claim within six months before or after Time 2 (1989
survey) defined as Group D 16 Respondents
29.0% fifteen years or more with the
organization
63.0% full-time/day shift employees
25.0% completed bachelor's degree
75.0% described as white female
31.0% were registered nurses
14.0% described as professional
Hospital 2 employees who filed stress-related medical
claims six; months before or after Time 2 (1989 survey)
defined as. Group D 14 Respondents
36.0%
78.0%
50.0%
50.0%
29.0%
36.0%
two to five years with the organization
full-time/day shift employees
white female
completed bachelor's degree
were registered nurses
described as professional
Hospital 1 Group C Employees who did not file claims
1989 601 Respondents
20.3%
25.0%
59.0%
66.0%
17.8%
18.0%
two to five years with the organization
completed a bachelor's degree
white female
full-time/day shift employees
were registered nurses
described as professional
Hospital 2 Group C Employees who did not file claims
1989 443 Respondents
19.4%
59.4%
22.8%
59.4%
22.6%
16.5%
five to ten years with the
organization
full-time/day shift employees
completed a bachelor's degree
white female
were registered nurses
described as licensed or technical
74


also counted in the second data collection if their
medical claim occurred six months before or after
the second survey (1989). This number was very
small. Employees who did not file claims and did
respond to both surveys were counted in the first
and second data collection analysis as part of the
total hospital MBI score. It was not possible to
separate the data unless different research
questions were asked. The research questions also
separated the data so that the results of the data
did not impact the findings of the particular
research question being examined. For example,
Group A was compared to Group B at Time 1 (1987) and
Time 2 (1989). Groups C of Hospital 1 was compared
to Group D at Time 2. The Group A and B data may
have been counted with Group C and D of Hospital 1
at Time 2 but the number would be very small.
Maslach Burnout Inventory
The standardized instrument used in this
study was the Maslach Burnout Inventory (MBI) as
revised by Golembiewski and Munzenrider (1981).
This twenty-five item instrument was devised to
measure hypothesized aspects of the burnout syndrome.
Appendix B contains the form of the MBI utilized in
this study. The MBI is designed to measure three
75


components that make up the burnout syndrome.
Emotional exhaustion, depersonalization, and a lack
of personal accomplishment are the subscales that
provide the information for the instrument.
According to Maslach and Jackson (1986) the
emotional exhaustion subscale assesses the feelings
of being emotionally over-extended and exhausted by
one's work. The depersonalization subscale measures
"an unfeeling and impersonal response towards
recipients of one's service, care, treatment, or
instruction." (p. 2). The personal accomplishment
subscale makes an assessment of the person's
feelings of accomplishment and work with people. A
high degree of burnout is reflected in the high
scores in the emotional exhaustion and depersonal-
lization subscales and then a low score on the
personal accomplishment subscale. A low degree of
burnout is reflected in low scores on the emotional
exhaustion and depersonalization subscaleand in
high scores on the personal accomplishment subscale.
Maslach and Jackson (1986) state that the
scores are not combined but considered separately.
Per the 1986 edition of the MBI manual, the
internal consistency was estimated by Cronbach's
coefficient alpha. The reliability coefficients
for the subscales were the following:
76


.90 for emotional exhaustion; .79 for
depersonalization; and .71 for personal
accomplishment. The standard error of measurement
for each subscale is as follows: 3.80 for emotional
exhaustion; 3.16 for depersonalization; 3.73 for
personal accomplishment.
Convergent validity was responded to by the
authors by the following mechanisms. The
individual's scores were correlated with behavioral
ratings made independently by a person who knows the
individual well. The MBI scores were correlated
with the presence of certain job characteristics
that were expected to contribute to the experience
of burnout. The authors also state in the manual
that the MBI scores "were correlated with measures
of various outcomes that had been hypothesized to be
related to burnout." Support for the reliability of
the MBI has been reported by Golembiewski,
Munzenrider, and Carter (1983), Belcastro, Gold, and
Hayes (1983), and Jackson (1985). The form of the
MBI included in the questionnaire used in the study
followed the approach of Golembiewski, Munzenrider,
and Carter (1983). The questionnaire had responses
to each question that asked the respondent to answer
to the degree to which the statement is like or
77


unlike you; This is different than the two
responses for intensity and frequency on the
original MBI. Golembiewski, Munzenrider, and Carter
(1983) suggested that this form of the MBI does not
distort Maslach's intent and the internal
reliability coefficients for the modified form are
almost the same as Maslach and Jackson's. The MBI
used in this study is better for this study than the
original due to the enhanced ability to interpret
the results according to Golembiewski and
Munzenrider, and Carter (1983).
Turnover, Demographic and other information
Collected
The turnover data collected for this study
was provided by the human resource department of the
hospital system that contained Hospital 1 and
Hospital 2. The turnover rates for 1987, 1988,
1989, and through September of 1990 were collected
for each hospital. Turnover rates for those
employees from the second data collection who had
filed medical claims (Group D) were from Hospital 1
and Hospital 2. These turnover rates were compared
to the four-year average turnover rate of the
hospital the employee belonged to. By doing this,
the analysis was able to examine the turnover rates
78


of employees who had filed stress-related medical
claims and those who had not.
The survey questionnaire included question-
naires pertaining to demographic information related
to the respondents. The hospital department where
the employee works, years of employment, employment
status and shift, highest level of education
completed, race and sex, and current organizational
position was collected. Medical claims data
provided the diagnosis of the medical claim, dates
of utilization, and charges for the services
rendered.
Analyses
The four research questions focused on in
this study are: (1) Is there a relationship between
burnout Scores and the incidence of medical claims
considered to be stress-related? (2) Will there be a
higher turnover rate in the organization in which
employees will be relocating than employees not
physically moving? (3) Will there be a difference in
the turnover rates of employees who file stress
related medical claims and those who do
not? and (4) Will burnout scores be higher for
employees relocating than employees not
relocating? All analyses were performed using the
SPSS-X Program on the University of Colorado at
79


Boulder VAX computer system. T-tests were computed
for the groups of employees who had filed claims and
those who had not. This was done for the first and
second data collections. T-tests were also utilized
to compare the differences in the means of the
Hospital 1 and Hospital 2 total organization burnout
scores at 1987 (Time 1) and 1989 (Time 2). All t-
tests were significant at the .05 level of
significance.
Analysis of the Four Research Questions
The first research question, "Is there a
significant relationship between burnout scores and
incidence of stress-related medical insurance
claims?" included MBI data gathered from the 1987
(Time 1) and 1989 (Time 2) surveys. Data from
employees from Hospital 1 (hospital relocating) who
responded, to both surveys provided one sample, Group
A, (n=200) for the first question. The Group A
sample was changed from an n of 200 to an n of 183
due to 17 employees being separated from the initial
200 and labeled Group B (employees not filing
claims). Group A (employees not filing claims) and
Group B (employees filing claims) MBI scores from
1987 (Time 1) and 1989 (Time 2) were analyzed.
Both samples were analyzed by using pooled
80


variance t-'tests at the .05 level of significance.
The t-tests were done using the means of the groups
filing claims and those who did not between the
years (1987 and 1989) the surveys were administered.
The addition of the subscales of the MBI scores of
Group A (employees not filing claims) and Group B
(employees filing claims) were also compared with
each other. The one score approach was done to
provide the study with another perspective of the
data collected.
A second data collection was undertaken for
research question one due to the small sample size
of employees who filed a claim (Group B) of the
first data collection. An attempt was made to
increase the data of employees who filed claims by
utilizing MBI scores of employees of both Hospital 1
and Hospital 2 who filed claims. Employee groups
who filed claims and were part of the second data
collection were labeled Hospital 1 Group D and
Hospital 2 Group D. Only 1989 (Time 2) data were
utilized. Medical claims information was also
limited to six months before and after the 1989
survey. The same procedures and coding process was
utilized for the second data collection including
the totaling of the subscales of the MBI for the
81


groups of employees who filed claims and the group
who had not.
The second research question, "Will there be
a higher turnover rate in the organization in which
the employees will be relocating to another location
than the rate of turnover in the organization not
physically moving?" was a comparison of turnover
data of each of the two hospitals over a four-year
period of time. Turnover rates were defined as the
percentage of employees of each hospital workforce
who left the employment of that specific hospital.
Hospital 1 (hospital relocating) was defined as the
hospital work force making the future physical move
and Hospital 2 (hospital not relocating) was defined
as the hospital not making the physical move.
The third research question, "Will there be
a difference in turnover rates between individuals
who file a claim for stress-related medical illness
and those who do not?" focuses on exploring possible
differences between those individuals who file a
medical claim for a stress related medical illnesses
and a behavioral outcome, i.e., turnover. Turnover
rates of employees who did file medical claims were
compared to employees who did not file medical
claims. Hospital 1 employees were compared to
82


Hospital 1 employees. The same process was used for
Hospital 2 employees.
The fourth and final research question, "Are
the burnout scores higher for employees physically
moving or relocating to another location in the
merger process than those employees not moving?"
Burnout scores of both hospitals were compared in
1987 (Time 1) and 1989 (Time 2). Pooled variance t-
tests at the .05 level of significance were utilized
to examine the difference between the means of the
burnout scores of employees located at each hospital.
The totaling of the hospital mean MBI subscale
scores, and burnout phase scores were examined for
possible differences between the two hospitals at
different points of time.
83


CHAPTER IV
FINDINGS
The merger impact models utilized in this
study proposed that the process of merging
organizations negatively affects employees of the
organizations merging. The research questions of
this study attempted to examine selected subjective
and objective aspects of a merger process. Employee
burnout scores were used to provide a perspective of
employees perceived stress in the organizations
studied. Medical claims information and turnover
rates were utilized to provide objective organiza-
tional measurements of employee stress. The impact
of one hospital being told that it would be
relocating staff and services in the future as part
of the merger process was analyzed to provide the
study with an event that could be specifically
connected to a result of merger activity.
MBI scores, turnover rates, and medical
claims information data were analyzed from
different perspectives so that a composite picture
of the impact mergers have on organizations could be
developed. The addition of time series data did
compare groups over time.


Research Question One
Research question one asked if there would
be a difference between the MBI subscale scores of
employees who filed stress-related claims and
employees who did not. The approach to the question
was to examine data from employees of Hospital 1
(hospital relocating) that responded to both surveys
(1987-Time 1 and 1989-Time 2). The groups were
separated by whether they filed a claim or did not
file a claim. The MBI subscales were analyzed, the
burnout phases were compared between the two groups,
and the subscale totals of the mean MBI scores were
totaled. The three approaches to the data were used
to provide as much information as available to the
study. The use of time series data enabled the
study to compare the groups at two distinct times.
After the first data collection, it was decided to
undertake the second data collection due to the
small sample size of employees who did file claims
between January, 1987, and December, 1989. The
second data collection included data from just the
1989 (Time 2) survey and from both Hospital 1 and
Hospital 2.
85


Results
After reviewing the first data collection
t-test differences of the MBI subscale scores and
the burnout phases, the results were analyzed.
Table 4-1 presents the results. The Time 2 (1989)
depersonalization subscale means of Group A
(employees not filing claims) and Group B (employees
filing claims) were significant at the .05 level of
significance. The two-tailed probability for those
two means were .029. The personal accomplishment
subscale scores at Time 1 (1987) for Group B
(employees filing claims) had a two-tailed
probability of .014. However, the majority of the
subscale scores were not significant at the .05
level of significance. Table 4-2 shows the
comparison of Group A at Time 1 (1987) with Group A
at Time 2 (1989). The same comparison has was done
for Group B. This analysis of each group at Time 1
compared to Time 2 (1989) was completed to see if
the specific groups changed over time. None of the
data reviewed was found significant at the .05 level
of significance. There was also interest in
determining if Group B employee scores showed
improvement. If Group B employees did file claims
and received medical attention, the impact of that
86


TABLE 4-1
Hospital 1 (hospital relocating)
Group A (employees not filing claims) and
Group B (employees filing a claim)
Compared at Time 1 (1987) and Time 2 (1989)
First Data Collection

MBI Subscale Sample Size Mean Standard Deviation Two-tailed Probability of Difference
Time 1 Emotional Exhaustion Group A 171 26.5029 9.948 .592
Group B 14 24.6429 12.304
Time 2 Emotional Exhaustion Group A 126 25.5852 9.733 .357
Group B 17 28.3529 11.683
Time 1 Depersonalization Group A 166 20.7711 7.532 .251
Group B 14 17.8571 8.857
Time 2 Depersonalization Group A 169 20.3195 8.107 .029*
Group B 17 15.6471 7.762
Time 1 Personal Accomplish- Group A 165 25.8182 7.407 .014*
ment Group B 13 21.6154 5.108
Time 2 Personal Accomplish- Group A 167 25.1916 7.442 .404
ment Group B 17 23.6471 7.079
Time 1 Burnout Phase Group A 154 4.8961 2.360 .051
Group B 13 3.3077 2.594
Time 2 Burnout Phase Group A 158 4.7342 2.577 .378
Group B 17 4.1176 2.690
* Significant at p <.05


TABLE 4-2
Hospital 1 (hospital relocating)
Group A (employees not filing claims) Time 1 (1987)
Compared to Group A Time 2 (1989) and
Group B (employees filing claims) Time 1 (1987)
Compared to Group B Time 2 (1989)
First Data Collection

Grout) A MBI Subscale Sample Size Mean Standard Two-Tailed Proba- Deviation bilitv of Difference
Time 1 Emotional Exhaustion n = 165 26.4727 10.056 . 198
Time 2 25.4485 9.885
Time 1 Depersonalization n = 156 20.7244 7.594 .482
Time 2 20.2308 8.180
Time 1 Personal Accomplishment n = 153 25.5163 7.287 .432
Time 2 24.9412 7.505
Time 1 Burnout Phase n = 139 4.8993 2.387 .303
Time 2 4.6547 2.595
Standard Two-Tailed Proba-
Grout) B MBI Subscale Sample Size Mean Deviation bilitv of difference
Time 1 Emotional Exhaustion n = 14 24.6429 12.364 .168
Time 2 28.7143 10.049
Time 1 Depersonalization n = 14 17.8571 8.857 .483
Time 2 16.6429 8.215
Time 1 Personal Accomplishment n = 13 21.6154 5.108 .208
Time 2 24.6923 7.543
Time 1 Burnout Phase n = 13 3.3077 2.594 . 141
Time 2 4.4615 2.787
Significant at p <.05.


medical attention was of interest. A positive
impact defined as improved burnout scores did not
occur at the .05 level of significance.
Table 4-2 also shows that the burnout phases
were not significantly different for the groups not
filing claims than those that did. This may
indicate that the level of burnout does not
necessarily lead to a stress-related medical claim.
After the results of the first data
collection, it was determined to collect more data
due to the small sample sizes available in the first
data collection. The second data collection
consisted of employees from Hospital 1 (hospital
relocating) and Hospital 2 (hospital not relocating).
These employees had responded to the Time 2 (1989)
survey. The group labeled Group C did not file a
claim. The group labeled Group D did file a claim
within six months of the second survey. The t-test
and analysis for different results are provided in
Table 4.3. This data supported the results of the
first data collection. A significant difference
between the means was not found at the .05 level of
significance for any of the subscales. The MBI
subscale and burnout phase means, standard
deviations, and two-tailed probability of difference
are provided in Table 4.3.
89