Citation
Life stress

Material Information

Title:
Life stress impact on genital herpes recurrences
Creator:
Hoon, Emily Franck
Publication Date:
Language:
English
Physical Description:
vii, 112 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Diseases ( jstor )
Genital herpes ( jstor )
Infections ( jstor )
Lesions ( jstor )
Life events ( jstor )
Locus of control ( jstor )
Psychological stress ( jstor )
Psychology ( jstor )
Questionnaires ( jstor )
Simplexvirus ( jstor )
Clinical and Health Psychology thesis Ph.D ( mesh )
Dissertations, Academic -- Clinical and Health Psychology -- UF ( mesh )
Herpes Genitalis ( mesh )
Life Change Events ( mesh )
Stress, Psychological ( mesh )
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph.D.)--University of Florida, 1986.
Bibliography:
Bibliography: leaves 99-110.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Emily Franck Hoon.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright [name of dissertation author]. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
023216545 ( ALEPH )
17865016 ( OCLC )
ACY6046 ( NOTIS )

Downloads

This item has the following downloads:


Full Text














LIFE STRESS: IMPACT ON
GENITAL HERPES RECURRENCES






BY






EMILY FRANCK HOON


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY


UNIVERSITY OF FLORIDA


1986




LIFE STRESS: IMPACT ON
GENITAL HERPES RECURRENCES
BY
EMILY FRANCK HOON
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1986


ACKNOWLEDGMENTS
This project involved the time, talent, effort, and cooperation
of many individuals. Because I appreciate their contributions and
want to recognize them and thank them, I address the following:
Ken Randpatient, enthusiastic, humorous, encouraging,
knowledgable, accessiblefriend: contributed freelyremembered with
warmth forever!
Jim Johnsonguidance, sense of organization, perspective,
calming presence, wisdoma role model deserving emulationmy sincere
thanks.
Randy Carter and his assistant statisticians who helped make
sense of voluminous computer printouts.
And to the rest of the best committee a Ph.D. candidate ever
had: Eileen Fennell, Rudy Vuchinich, and Nancy Norvell.
Wendy and Robin Hoonyoung people grown independent, before my
eyes, out of necessity while coping with the vagaries of their Mom's
graduate school scheduleI love both of you, and I thank you for your
help and cooperation.
Ernest Franck, my dad, who has always encouraged me to follow my
curiosity and to accept challenges.
The late Emily Franck, my rnom, who helped me to appreciate the
value of social relationships.


Pete Hoon, my ex-husband, with whom I shared my early research
experiences and with whom I continue to enjoy a satisfying
collaborative relationship as a professional and coparent.
Rob Martin, my present husband, who helped me in uncountable ways
to tie up loose ends and who provides inspiration for getting the most
out of work time and play time.
Subjectscourageously sharing personal information to advance
knowledge regarding herpes.
And the following cheerful helpers: research assistants,
laboratory personnel, and Clinical Research Center staff.
i i i


TABLE OF CONTENTS
Page
ACKNOWLEDGMENTS ii
ABSTRACT vi
CHAPTERS
ONE LITERATURE REVIEW 1
Stress and Illness ...1
Moderator Variables 4
Social Support 4
Locus of Control 7
Arousal Seeking 9
Methodological Issues 12
Retrospective Design 12
Measures of Illness 14
Confounding of Events and Outcome Measures 15
Value and Impact of Events 16
Minor Events 17
Recurrent Genital Herpes as a Model 18
Clinical Nature of Herpes Simplex Virus 18
Recurrences 19
Present Lack of Cure 21
Laboratory Documentation 21
Endogenous Pathogen 22
Stress and Herpes 23
Anecdotal Reports 23
Survey Research 25
Retrospective Research 26
Prospective Research 28
Summary 31
Theoretical Framework 33
Perception of Stressful Events 33
Proposed Physiological Mechanisms 36
Present Investigation 44
Hypotheses 45
TWO METHODS 48
Research Design 48
Sample Selection 49
i v


Measures 50
Stress Measures 50
Life Experiences Survey (LES) 50
Hassles Scale (HS) 51
Moderator Variables 52
Locus of Control (LC) 52
Arousal Seeking (AS) 53
Social Support Questionnaire (SSQ) 54
Physiological Measures 55
Screening procedure: HSV antibody 55
Dependent variable: Virus isolation and
transport 55
Procedure 56
Initial Contact 56
Second Contact 57
Continuing Contact 53
Final Contact 58
THREE RESULTS 59
Sample Characteristics 59
Undesirable Life Events 60
Presumed Moderator Variables 64
Definition of Recurrence 65
Stress/Recurrence Relationship 67
Influence of Moderator Variables 68
The Nature of the Life Stress/Recurrence Relationship 75
FOUR DISCUSSION 78
APPENDICES
A STRUCTURED INTERVIEW 91
3 SUBJECTS' TOTAL SCORES ON STRESS, MODERATOR, AND
RECURRENCE VARIABLES 94
C SPEARMAN CORRELATION COEFFICIENTS AND SIGNIFICANCE
LEVELS FOR REPORTED STRESS FREQUENCY AND RECURRENCE
DEFINITIONS 98
REFERENCES 99
BIOGRAPHICAL SKETCH Ill
v


Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
LIFE STRESS: IMPACT ON
GENITAL HERPES RECURRENCES
BY
EMILY FRANCK HOON
May, 1986
Chairman: James H. Johnson
Major Department: Clinical Psychology
The present study of the effects of major and minor life events
on recurrences of genital herpes addressed the need for prospective
research with objective indices of illness (i.e., viral culture) and
provided empirical information regarding a stress/recurrence link.
For 6 months, 122 HSV seropositive subjects monitored stress and
had all recurrent lesions cultured. Subjects were administered the
Life Experiences Survey (LES) upon enrollment regarding major events
experienced over the past 6 months. This information was analyzed to
determine the extent of a prospective relationship to subsequent
recurrences. Thereafter the LES was administered monthly with the
Hassles Scale (HS) which concerns minor stressors. To identify
personal characteristics which might be related to frequent
recurrences under conditions of stress, the following presumed
moderator variables were examined: social support, locus of control
orientation, and arousal seeking tendency.
vi


In accordance with the first hypothesis, correlational procedures
provided evidence for a positive association between the frequency of
major negative life events and the number of HSV recurrences during
the study. This relationship was significant for concurrent LES
scores, but not when initial LES scores were considered in a
prospective manner to subsequent recurrences.
A logistic regression procedure using all subjects provided
evidence for a moderating influence of social support satisfaction,
the nature of which was contrary to expectation. A significant
positive stress/recurrence relationship was found only at high levels
of satisfaction. As hypothesized, information from recurring subjects
(jr=49) submitted to a regression analysis revealed a significant
positive stress/recurrence relationship only for externally oriented
individuals. Neither arousal seeking nor hassles were significantly
related to recurrences in any analysis.
Within-subject analyses were performed to examine the nature of
the concurrent stress/recurrence relationship. Stress was not found
to be significantly elevated prior to recurrences. This suggests
either a shorter latency exists between stress and recurrence onset
than could be detected; using monthly evaluations or a reciprocal
relationship exists in which stress may cause or be caused by
vii
recurrences.


CHAPTER ONE
LITERATURE REVIEW
Stress and Illness
The notion of a link between psychosocial factors and illness is
not new. Physicians have prescribed environmental and behavioral
interventions for years based upon their clinical experience and
intuition. What is new is a flurry of activity designed to document
and clarify this relationship.
With regard to life stress as a psychosocial factor in the
etiology of illness, Holmes and Rahe (1967) can be credited with
initiating a new wave of research with the development of their
measure, the Schedule of Recent Experience (SRE). This was the first
attempt to quantify, using a simple, 43-item questionnaire format, the
stressful life events believed to heighten vulnerability to illness.
The SRE and various modifications of this measure have been
widely used to provide empirical evidence for the relationship between
life stress (defined ip terms of major life changes) and illness.
Consistent but modest correlations have been confirmed with a variety
of illnesses and illness-related measures in the past two decades.
Elevated life stress has been shown to be associated with higher
reported illness rates among many groups (Holmes & Masuda, 1974; Marx,
Garrity, & Bowers, 1975; Rahe, Mahan, & Arthur, 1970) the seriousness
of illness (Wyler, Masuda, & Holmes, 1971), athletic injuries
1


3
through increased susceptibility to a range of health-related
problems.
In addition to studying the relationship between life stress and
illness onset, other studies have focused on the relationship between
life stress and fluctuations of health status in individuals with
chronic illness. Children with a mixed group of chronic conditions
(e.g., diabetes, asthma, blindness) who attended a 3-week summer camp
were observed by Bedell, Giordani, Armour, Tavormina, and Boll
(1977). A relationship between self-reported life change and
frequency of illness-related symptomatic episodes was found.
Similarly, asthmatic individuals monitored prospectively were found to
require more medication under conditions of increased life stress
(de Araujo, Van Arsdel, Holmes, & Dudley, 1973). Diabetic control has
also been investigated. Bradley (1979) found life stress to be
related to increases in blood sugar, the need for clinic visits, and
necessary prescription changes, particularly for insulin-dependent
diabetics. Similarly, insulin-dependent diabetics were found to be
more vulnerable to life stress than age- and sex-matched noninsulin
dependent diabetics and matched nondiabetic controls (Linn, Linn,
Skyler, <$ Jensen, 1983). Insulin-dependent diabetics perceived more
stress and showed poorer metabolic control despite similar compliance
with the medical regimen. They also demonstrated decreased immune
responsivity (Linn et al., 1983). Jacobson, Rand, and Hauser (1985)
studied the impact of life events on the long term complications of a
group of insulin-dependent diabetics. A small but significant
relationship was found between negative life events and glycemic


4
control for all subjects. In the subgroup of diabetics with
proliferative retinopathy of recent onset, a stronger relationship was
found between lack of glycemic control and the occurrence of negative
life events (Jacobson, Rand, & Hauser, 1985).
Moderator Variables
Despite the consistent finding of statistically significant
correlations, the correlation coefficients are of low magnitude,
typically less than .30, which suggests that life events measures
typically can only account for 9% or so of the variance in illness
measures (Rabkin & Struening, 1976). To improve upon this situation,
the life events measures have been administered in conjunction with
other measures believed to assess moderators of life stress.
Constitutional predisposition (Kobasa, Maddi, & Courington, 1981),
social support (Nuckolls et al., 1972; Schaefer, Coyne, & Lazarus,
1981), locus of control (Toves, Schill, & Ramanaiah, 1981), hardiness
(Kobasa, 1979), physical fitness (Roth & Holmes, 1985), and a tendency
to engage in arousing situations (Johnson, Sarason, & Siegel, 1979)
have been shown to moderate the effects of life stress.
*
Social Support
A frequently researched moderator variable has been social
support. This refers to the degree to which individuals have access
to social resources. Resources depend upon the relationships people
have to spouse, family, friends, neighbors, community groups, and
social institutions (Johnson & Sarason, 1984). Social support has


5
been measured in a variety of ways, from simply confirming the
availability of a confidante (Brown, Bhrolchain, & Harris, 1975) to
administering a 48-item inventory with four subscales to evaluate
esteem support, instrumental support, social companionship, and
informational support (Cohen & Hoberman, 1983).
In a recent comprehensive review of over 40 research articles,
Cohen and Wills (1985) found evidence consistent with both an overall
beneficial effect of support on well-being and a buffering
(interactive) role in relation to life stress. In the former, health
benefits are perceived to derive directly from a supportive network.
In the latter, social support is perceived to play a moderating role,
to be related to well-being only for persons experiencing the adverse
effects of stressful events. That is, in the face of major life
stress, an adequate social support system may serve to protect the
stressed individual from a pathologic outcome. Authors point out that
social support may simultaneously exert both a direct and a buffering
effect (Cohen & Wills, 1985).
Social support was found to exert buffering or positive main
effects on most of the psychological distress variables reviewed, such
as depression, loneliness, and anxiety. In fact, in one investigation
(Lin, Ensel, Simeone, & Kuo, 1979), the contribution of social support
to predicting psychiatric symptoms was more important than stressful
life events. However, findings regarding physical symptomatology have
been more variable. A clear, consistent link between social support
and decreased mortality has been demonstrated (Berkman & Syme, 1979),
but studies utilizing other health outcomes are less clear. They are


mixed when the health outcome is self-reported symptomatology. When
more stringent health measures have been used such as clinical
diagnosis and alterations in physiological functioning, a beneficial
role for social support has not always been consistently demonstrated
(Cohen & Wills, 1985).
A few examples will be given to illustrate the diversity of
outcomes regarding social support. Evidence consistent with a stress
buffering hypothesis comes from an early study of pregnant women by
Nuckolls et al. (1972). Those experiencing high levels of stress and
low levels of social support experienced almost three times as many
pregnancy and birth complications as their similarly highly stressed
peers who had high levels of social support. Similarly, adult
asthmatics in a high life stress-low social support group required
significantly more medication to control symptoms than their peers
with either less stress or more social support (de Araujo et al.,
1973). These examples demonstrate a protective, positive role for
social support with respect to certain physical outcomes.
In spite of the intuitive appeal of a positive effect of social
support, and in contrast to the above studies, Norbeck and Tilden
(1983) found no evidence for either a main effect or buffering role
with respect to pregnancy complications. They did, however, find a
positive main effect of emotional support on psychological
symptomatology (Norbeck & Tilden, 1983). New Zealand researchers
(Graves & Graves, 1985) failed to find the hypothesized buffering
effect of social support on self-reported symptoms of illness in three
distinct ethnic groups. In fact, of the correlations designated large


7
enough to be of clinical significance, all were in direct opposition
to the buffering hypothesis. It is suggested that the obligatory
reciprocity of dense social networks may be less helpful under
conditions requiring adaptation to life change (Graves & Graves,
1985). An interesting cyclical pattern of stress and social support
was demonstrated by Canadian investigators in a longitudinal study
(McFarlane, Norman, Streiner, & Roy, 1983). They found that help from
social support networks reduced exposure to stressful events.
However, increases in stressful events led to a reduction in perceived
he1p. No association was found between social support and physical
symptom reporting or physician visits (McFarlane et al., 1933).
Obviously, the relationship between social support, stress and
health is a complex one. Critics suggest the use of longitudinal
designs (Cohen & Wills, 1985; Thoits, 1982) and the measurement of
satisfaction (perceived quality) with available support as well as the
amount (quantity) of social support available (Sarason, Levine, &
Sarason, 1982) to further unravel the role of social support with
respect to major life events and health status.
Locus of Control

Another measurable variable that has shown promise as a moderator
of life stress is an individual's perception of the controllability of
situations. People vary in the degree to which they attribute the
responsibility for events to themselves or to fate. It is reasonable
to assume that individuals who perceive themselves as having little or
no control would feel more threatened in the face of undesirable life


3
events than those who feel capable of influencing outcomes. A more
intense adverse reaction could contribute to a more severe
physiological response to life stress.
The Rotter (1966) Locus of Control Scale measures the degree to
which individuals perceive themselves as having control over naturally
occurring life events. It has provided a tool for investigation of
perception of control, life stress, and psychological and physical
outcomes. Individuals are classified as internals on the scale if
they believe in personal responsibility and externals if they
attribute the control of events to chance. Johnson and Sarason (1978)
administered the Locus of Control Scale along with a major life events
scale and measures of depression and anxiety to college students.
They found negative life changes to be significantly related to both
trait anxiety and depression for externals, but not for internals. A
similar relationship was found for males in a study of life stress,
locus of control, and health status as measured by the Cornell Medical
Index (Toves et al., 1981). Negative life change was related to
health status only for external males. Females demonstrated a
relationship between negative events and illness regardless of
perception of control (Toves et al., 1981). In further support of the
notion of perceived control as a moderator of life stress, Kobasa
(1979) demonstrated that individuals with an internal locus of control
may be protected from adverse physical consequences of major life
change. Compared to business executives with high stress/high illness
scores, those with high stress/low illness indicated a more internal
locus of control. In this investigation, perception of internal


9
control emerged as one component of a constellation of factors named
"hardiness" (Kobasa, 1979).
The above studies are all retrospective in nature, and it could
be argued that individuals who have experienced recent illness would
see themselves as victimized and thus tend to respond to the locus of
control scale as externals because of their poor health status. To
strengthen the causal inference that hardiness (which includes an
internal locus of control orientation), Kobasa (1981) replicated her
findings in a prospective design suggesting that an internal locus of
control is related to resistance to illness under conditions of
stress. Manuck, Hinrichsen, and Ross (1975) followed college students
for 6 months after assessing locus of control orientation and life
stress. Although they found no difference in illness-related
treatment-seeking behavior between internals and externals under high-
stress conditions, more low-stressed externals sought treatment than
low-stressed internals (Manuck et al., 1975). Passer and Seese
(1983), however, failed to prospectively demonstrate a significant
moderating effect of perception of control with athletic injury as the
dependent variable. These findings suggest that locus of control
merits further investigation as a moderator of the effects of major
life stress.
Arousal Seeking
Individuals differ in the degree to which they seek out or
attempt to avoid arousing situations. Inventories have been devised
to evaluate this tendency enabling researchers to investigate this


10
characteristic as a moderator of life stress (Mehrabian, 1978;
Mehrabian & Russell, 1974; Zuckerman, 1971). It is hypothesized that
those with a low optimal level of arousal, who tend to avoid
stimulation (low arousal seekers), would be more likely to respond
negatively to undesirable life events than those who have a high
optimal level of arousal and thus seek out stimulation (high arousal
seekers). High arousal seekers are presumed to be better able to deal
with the increased arousal brought about by the experiencing of life
changes (Johnson & Sarason, 1984).
This prediction has been addressed by a few researchers. Smith,
Johnson, and Sarason (1978) found negative life change to be
significantly related to discomfort scores on the Psychological
Screening Inventory for only the college students low on a sensation
seeking measure. Extending these findings using a different measure
of arousal seeking and measures of anxiety, depression, and hostility,
Johnson et al. (1979) found negative change to be related to anxiety
and hostility only for low arousal seeking subjects. At least one
researcher (Cohen, 1982) has failed to replicate the above results.
These preliminary findings with regard to psychological outcomes
suggest that a moderating effect of arousal seeking tendency on life
stress with regard to certain physical outcomes may also be worth
investigation. Individuals low on the arousal-seeking dimension may
be much more likely to experience physiological reactivity in the face
of life stress than others, which could adversely affect general
heal th.


11
In summary, as a means of investigating the variability in
individual susceptibility to health status changes in the face of
undesirable life stress, the investigation of the roles of moderator
variables appears to be warranted. The potential influence of social
support, locus of control, and arousal seeking tendency are of
particular interest based on findings to date.
With the exception of the Kobasa investigations (Kobasa, 1979;
Kobasa et al., 1981), moderator variables have been evaluated singly
regarding their potential to alleviate or exacerbate the negative
effects of life stress. More information is needed on how these
variables interact or combine in an additive fashion to influence
health status in the face of life stress. Multiple regression
techniques provide an analytic tool appropriate for clarifying the
relative importance of multiple potential moderator variables.
Increased documentation of significant associations between
stressful life events and the onset or course of physical illness has
changed the degree of complexity of the empirical questions. Instead
of asking vf stress is related to a particular illness, the questions
to answer now are for which individuals, under what conditions, and to
what extent are life events related to this disease? Answering more
complex questions requires more complex research designs (Johnson &
Sarason, 1984). In the next section, some of the methodological
shortcomings of present research will be discussed. Addressing these
issues will enable us to be more specific about our research questions
and to further clarify the nature of the stress/i11 ness relationship.


12
Methodological Issues
Methodological issues limit interpretations and findings in this
area of research. These include retrospective bias, inadequate
indices of illness, failure to discriminate between desirable and
undesirable events, and failure to account for minor stressful events.
Retrospective Design
The commonly utilized retrospective design relies on recall of
events and illnesses. Findings based on memory may reflect
individuals' expectations regarding stress and illness, their need to
justify illness (Rabkin & Struening, 1976), or their tendency to
report negatives due to feelings of depression (Johnson & Sarason,
1984). Retrospective reporting is also subject to forgetting as
demonstrated by Monroe (1982a). He investigated memory of major life
events utilizing prospectively collected information as a baseline to
indirectly estimate the most recent retrospectively reported four-
month period. Underreporting of events was estimated in this way to
be as high as 61$, with desirable events relatively more susceptible
to distortion with passing time than undesirable events (Monroe,
1982a). Funch and Marshall (1984) examined the rate of fall-off in
event reporting over a retrospective 30-month period. Using estimates
based on 12- and 18-month recall periods, they concluded that fall-off
is most rapid in the first 12 months, approximately 5% per month. It
then tends to level off. Fall-off was related to type of event in
their research, the more salient events showing the least fall-off.
Respondent variables such as income, education, and marriage were


13
related to the tendency to remember disruptive events. That is,
subjects with fewer resources were most likely to underreport
stressful events (Funch & Marshall, 1984). In a review of several
articles which address the issue of event underreporting, Paykel
(1983) found consistency in estimates of 4-5% loss per month in
retrospectively self-reported life events, with a lower rate (1-3%)
for event reporting by interview. To avoid retrospective bias due to
forgetting, frequent short-time intervals for reporting have been
recommended (Cleary, 1980; Monroe, 1982a), with interview techniques
used whenever possible (Paykel, 1983).
Prospective studies avoid this kind of distortion. Jenkins,
Hurst, and Rose (1979) reviewed studies on cardiac and cancer patients
finding prospective studies which contradicted or diminished the
significance of earlier findings of retrospective investigations.
Unfortunately, good prospective studies of the onset of illness are
very cumbersome and costly to implement. Many people must be
investigated who may never get the disease or diseases in question
during the course of study. For example, of 1400 cadets studied for
four years, only 194 became infected with infectious mononucleosis
during the investigatipn and, of these, only 48 developed clinical
symptomatology (Kasl, Evans, James, & Niederman, 1979). In another
example, 4,486 widowers were followed for 9 years to conclude that
there is a 40% increase in mortality within the first 6 months of
bereavement (Parkes, Benjamin, & Fitzgerald, 1969). Prospective
investigations of the effects of life stress on the clinical course
and outcome of disease are more easily and economically


14
accomplished. These involve prospectively following already
identified patients with respect to life stress along with
fluctuations in their health status regarding the illness in
question. The data from all subjects remain important for analysis,
not just those who happen to die or become infected with a particular
disease.
Measures of Illness
Another weakness of research in this area has been the failure to
employ strict measures of illness (Johnson & Sarason, 1984). Self-
reports of symptoms, the seeking of medical treatment, and medication
needs may not reflect the existence or severity of a physical
disorder. Additionally, high life stress may increase the tendency to
seek help for minor illness (Mechanic, 1975) or lower an individual's
threshold of complaint and tendency to adopt the sick role (Minter &
Kimball, 1978). Harney and Brigham (1984) experimentally demonstrated
that an individual's tendency to tolerate discomfort (cold pressor,
loud noise) is related to the amount of recent life change
experienced. This suggests that high life change individuals may be
more likely to seek aid than others for identical symptoms (Harney &
t
Brigham, 1984). One way to avoid these confounding factors is to
employ more rigid measures of illness. Rather than reliance on
illness behavior, medical records, or self-report, more objective
indices are suggested such as x-ray and other clinical and laboratory
data which provide biological evidence of the presence or severity of
disease (Minter & Kimball, 1978).


15
Confounding of Events and Outcome Measures
A concern of many in this area of research in the possible
confounding of independent and dependent variables in life stress
measures (Johnson & Sarason, 1984; Monroe, 1982b). Many items which
are considered to be stressful events may also be consequences of
illness (e.g., sexual problems, major personal injury, changes in
eating or sleeping habits). An argument can be made that such items
should be removed or analyzed separately from the domain of major life
events to avoid artificially inflating the stress scores of
individuals with illness (Dohrenwend & Dohrenwend, 1974). However,
such experiences may themselves represent significant sources of
stress. To ignore them in a consideration of the total stress of an
individual may be to neglect important information and result in a
loss of sensitivity in the life stress measure. In an investigation
of etiological factors associated with a particular disease, a
division of items into three subgroups for analysis is recommended:
(a) events that may be confounded with psychiatric condition, (b)
events consisting of physical illness and injury, and (c) events which
are independent of either an individual's health or psychiatric
I
condition (Dohrenwend & Dohrenwend, 1974; Monroe, 1932b). When the
major purpose of the investigation is to predict illness from stress
levels, it is reasonable to include all items as reflective of the
total stress experienced (Johnson & Sarason, 1984). This would be
particularly appropriate when the illness dependent variable is
specific and objective (e.g., laboratory documentation of the presence


16
of a disease organism). Confounding problems are more likely when
general, subjective measures of illness serve as the dependent
variable (e.g., reported symptoms, treatment-seeking behavior).
Value and Impact of Events
Other methodological considerations include the value (negative
or positive) and impact (weighting) of events. Until recently,
investigators have failed to discriminate between desirable and
undesirable events. In those investigations in which a distinction
has been made, negative change has proven to be more predictive of
personal maladjustment (Mueller, Edwards, & Yarvis, 1977; Vinokur &
Selzer, 1975) and physical illness (Hotaling, Atwell, & Linsky, 1978;
Johnson & Sarason, 1984; Sheehan, O'Donnell, Fitzgerald, Harvey, &
Ward, 1979).
Different methods have been explored for weighting the impact of
stressful events to obtain a more sensitive measure of individual or
consensus-derived perception of the stressfulness of particular life
events. However, raw event totals and weighted event totals are
typically highly correlated with each other. Zimmerman (1983) found
an average correlation of .94 in a review of 19 studies. Since the
I
differential weighting procedures are functionally equivalent to
unweighted procedures in group research, it is parsimonious to employ
the simplest method: a simple count of events (Johnson & Sarason,
1984; Monroe, 1982b; Skinner & Lei, 1980; Zimmerman, 1983). This does
not deny the clinical relevance of subjective weighting procedures for


17
understanding particular individuals. It simply acknowledges that
such refinements are unnecessary and redundant for group research.
Minor Events
The final methodological issue relates to the consideration of
minor as well as major life events. Because major life events per se
do not represent all stressful events, the Hassles Scale (HS) has been
developed (Kanner, Coyne, Schaefer, & Lazarus, 1981). In contrast to
major life events such as divorce, job loss, and pregnancy, the HS
deals with minor stresses, annoying practical problems such as making
mistakes, being bored, and having to wait. Already there is some
suggestion that hassle scores may be more strongly related to a
general health measure than life events scores (DeLongis, Coyne,
Dakof, Folkman, & Lazarus, 1982).
In order to account for a greater proportion of the variance in
the relationship between stress and illness, the present study was
designed to address some of the existing methodological limitations.
To avoid retrospective bias, a prospective research design with
monthly concurrent assessments of stressors was planned. The Life
Experiences Survey (LES) was chosen to measure major life stress
because it distinguishes between negative (undesirable) and positive
(desirable) events (Sarason, Johnson, & Siegel, 1978). The Hassles
Scale (HS) was utilized to evaluate minor stressful events (Kanner et
al., 1981). Additionally, three hypothesized moderator variables were
measured to evaluate their effectiveness in combination in


18
ameliorating the effects of stress. And, finally, a population was
chosen for which an objective index of illness could be obtained.
Recurrent Genital Herpes as a Model
The choice to study individuals with recurrent genital herpes
virus in this investigation was based on the nature of the illness,
the need for information concerning the precipitants of reactivations
of the virus, the ease of obtaining an objective measure of the
disease, and the availability of a motivated population of subjects.
Given these considerations, an investigation of herpes simplex virus
would seem to be a fruitful means of studying the relationship between
life stress and fluctuating health status in chronic illness.
Clinical Nature of Herpes Simplex Virus
Herpes simplex virus (HSV) types 1 and 2 are two of several
herpes viruses which affect humans and share a unique capacity to
produce lifelong latent infection despite host development of antibody
and cellular immunity. Serious complications can occur including
ocular infection, encephalitis, and neonatal infection. Fortunately,
these are rare and can usually be prevented. There also appears to be
an association between genital herpes and cervical cancer, but so far,
direct evidence of a causal link is lacking (Rand, 1982).
Neither serious complications nor significant physical
consequences are responsible for the widespread fear and attention
elicited by the disease. There is evidence that the incidence of
genital herpes is rising rapidly. Consultations to clinicians for


19
genital herpes have increased almost nine-fold in the past 20 years
(Mertz & Corey, 1984). Its incurability, transmissabi1ity, rising
incidence, and sexual nature create the concern. This can only be
understood in light of information regarding the natural course of the
disease.
Herpes is almost always transmitted by intimate contact between
infected and noninfected skin surfaces, particularly the mucous
membranes of the oral cavity and genitalia. Once inside, herpes takes
over the protein-producing apparatus of the host cell and reproduces
itself until it is brought under control by the immune system. During
an initial (primary) infection, this process produces multiple,
painful blisters or ulcerative lesions accompanied by fever, swollen
lymph nodes, and malaise. The lesions last an average of 19 days
(Mertz & Corey, 1984).
Within hours of initial exposure, the virus retreats via nerve
fibers to a safe hiding place in the trigeminal ganglia (oral herpes)
or the sacral ganglia (genital herpes) which is located near the
spinal cord. It is this capacity of the virus to reside latently
without destroying the ganglion and to periodically reactivate to
cause recurrent lesions which frustrates patients, doctors, and
researchers.
Recurrences
Almost all patients experience recurrences, from one in a
lifetime to two or three a month. The risk of recurrence when the
infectious agent is HSV-2 is higher than with HSV-1 regardless of the


20
site of the infection. Otherwise, the clinical manifestation of HSV-1
and HSV-2 is identical (Mertz & Corey, 1984). A diverse group of
factors are associated with reactivation of HSV including fever
(Keddie, Reeves, & Epstein, 1941), local irritation or tissue injury
(Blyth & Hill, 1984; Wickett, 1982), lack of sleep, overexertion, and
emotional stress (Bierman, 1982). Although there is widespread
acceptance of a stress/recurrence relationship, the evidence for it is
weak. The notion is based on anecdotal reports and a handful of
retrospective studies. The further clarification of the nature of
this relationship is important for proper disease management.
The physical consequences of recurrences are minimal. The
lesions cause discomfort but are rarely severe. They may last up to 2
weeks, but are sometimes fleeting, lasting a day or less. The patient
can usually carry out normal daily activities during recurrent
episodes. Between recurrences patients are completely well. However,
patients are highly infectious during a recurrent episode and may
transmit the disease even when asymptomatic (Mertz & Corey, 1984).
Patients find the periodic imposed abstinence from sexual activity to
be disruptive and to result in decreased sexual spontaneity. Female
patients have concerns about their reproductive capability. Single
individuals must deal with disclosure to potential partners, causing
embarrassment, shame, and fear of rejection (Luby & Klinge, 1985).
Unfortunately, because of its sexual nature, biased reporting
regarding genital herpes in the media has contributed to popular panic
about the disease, stigmatizing patients, and exacerbating the
emotional turmoil involved (Bierman, 1985).


21
Present Lack of Cure
At the present time, no cure is available for the disease. It
has recently been demonstrated that oral administration of daily doses
of acyclovir suppresses reactivation of the virus in patients with
frequently recurring genital herpes (Douglas et al., 1984; Straus et
al., 1984). However, when the prophylactic therapy ceases,
recurrences resume at the prior rate. And, unfortunately, the expense
of the drug and as yet unknown long term side effects prohibit its use
by many. Until a cure can be found, management of the disease must
consist of counseling patients to prevent transmission by practicing
sexual abstinence during recurrences and preventing recurrences.
Prevention of frequent recurrences with acyclovir is an option for
some. With better knowledge of the precipitants of frequent
recurrences of genital herpes, behavioral counseling for drug-free
prevention for all patients will improve. Thus, there is a compelling
need to learn more about the psychological and behavioral factors
associated with recurrences.
Laboratory Documentation
It is possible to obtain objective evidence of the presence of
herpes simplex virus by laboratory methods. The virus may be
cultivated in tissue culture, viral particles may be detected by
electron microscopy, or immunologic methods may be used to detect
viral antigen or cytopathologic change. The isolation of the virus in
tissue culture has been shown to be the most sensitive measure. The
virus has been successfully isolated in 82% of ulcerative lesions in


22
primary genital herpes episodes and 42 of ulcerative lesions in
recurrent episodes. The decrease in percent viral confirmation in
recurrent episodes as opposed to primary can be attributed to a
shorter duration of viral shedding in the former (Corey & Holmes,
1983). The fact that objective evidence of the presence of the
disease is possible also contributed to the choice to study this
particular group of individuals in a stress/illness paradigm.
Endogenous Pathogen
A final factor regarding recurrent genital herpes is that
extraneous exposure to an unknown dosage of an exogenous pathogen was
not an issue. Stress would be most important as a contributing factor
in the etiology of a disease when the host/microorganism relationship
is in delicate balance (Plaut & Friedman, 1981). Uncontrolled,
extraneous dosages of a pathogen may be so high that disease is
inevitable, or too low to cause disease under any conditions.
Likewise, the more virulent the microorganism, the lower the dose
necessary to introduce infection. Recurrent herpes lesions depend
upon an endogenous source of infection which remains stable within an
individual unless reinfection occurs from an outside source.
Genital herpes, therefore, provides a good model for the study of
stress/illness relationships. It is a chronic, recurring disease
which causes emotional turmoil for patients and their sexual partners
despite minimal physical consequences. Between recurrent episodes,
patients are unaffected by disease presence. Patients are thus
motivated to avoid recurrences and willing to participate in research


23
to learn more about precipitants of herpes reactivations. Viral
isolation is possible by laboratory techniques providing an objective
measure of the presence of the disease, and extraneous exposure to
pathogens of unknown dose and virulence is not a concern. Patients,
physicians, and individuals who have never had genital herpes will all
benefit from improved understanding of the disease.
Stress and Herpes
In the following section, investigations which have attempted to
clarify the relationship between stress and herpes will be
presented. Herpes simplex virus-type 2 most commonly affects the
genital area, but HSV-1 may also do so. Because they are clinically
indistinguishable (Rand, 1982), research including either HSV-1 or
HSV-2 will be reviewed.
Anecdotal Reports
Schneck (1947) reported the case of a male who could predict
HSV-1 recurrences which occurred within 24 hours of emotional
stress. These episodes were most likely to occur when appropriate
outlets for hostile feelings were unavailable. According to the
account, improved management of hostility resulted in decreased
recurrences.
Blank and Brody (1950) provided psychoanalysis for 10 patients
with recurrent oral herpes. Patients were seen weekly for 2 to 50
hourly sessions. Nine of the ten patients were described as passive,
anxious to please, dependent, immature, and hysterical. In two cases,


cancellation of an appointment by the therapist and the resultant
feeling of rejection is said to have precipitated a herpes
recurrence. From their experiences with this group of patients,
authors claim support for the use of psychotherapy in decreasing the
frequency of recurrences (Blank & Brody, 1950).
There have been reports of an association between hypnotic
suggestion and recurrent herpes episodes. Herlig and Hoff (cited in
Janicki, 1971) successfully induced oral herpes recurrences in three
female patients by reminding them of unpleasant emotional situations
and itching sensations while under hypnosis. Recurrent lesions
appeared in 1 or 2 days in all three subjects. Ullman (1947)
similarly induced an oral herpes episode in a male patient by giving
hypnotic suggestions that herpes blisters would form and that he
appeared debilitated. Within 24 hours the man had blisters on his
lower right lip which appeared to be oral herpes. Hypnosis has also
been reported to have a positive effect on the recurrence rate of
HSV-1 (McDowell, 1959).
Many supposed "cures" of herpes have been attributed to placebo
effects, or positive expectancy. One investigator reports favorable
reactions in 50% of herpes patients who received injections of steril
water (Kern, 1979). Hamilton (1980) reports even more dramatic
placebo results. In a double-blind placebo-controlled trial of ether
as a treatment for herpes, 75% of the patients on placebo experienced
improvement of herpes symptomatology (Hamilton, 1980, p. 54).
Anecdotal reports of an association between hypnosis or
expectancy and episodes of recurrent herpes are suggestive, but


25
inconclusive. Failures to induce herpes blisters through suggestion
are unreported. However, the existence of a link between emotional or
cognitive factors and recurrent herpes as can be hypothesized from
these reports is worthy of more systematic investigation.
Survey Research
Results of epidemiological survey research also suggest a link
between emotional factors and recurrences. In a national survey of
3,148 individuals with genital herpes, 83% of the respondents
acknowledged stress as a factor in recurrences ("Help membership HSV
survey," 1981). The subjects attributed the following ill effects to
having contracted herpes: periodic depressions (84%), sense of
isolation (70%), conscious avoidance of intimacy (53%), cessation of
sexual activity (10%), diminished sex drive and/or impotence (35%),
suicidal feelings (25%), dissolution of a long term marriage or
relationship (18%), rejection by a potential sex partner (21%), and
loss of self esteem (40%). Herpes is cited as both a cause and effect
of stress by these participants who were recruited through the
newsletter of a support network for individuals with genital herpes.
Two-thirds of the sample reported experiencing more than five
recurrences in 1 year. Although the sample may not be representative
of all genital herpes patients, it does reflect the perceptions of a
substantial number of patients. Another, possibly more representative
sampling of 825 patients, 53% of whom believed they were cured of
genital herpes, yielded a similar finding. In this group, emotional


26
stress emerged as the most important self-reported mechanism for
triggering recurrences, cited by 86% of the sample (Bierman, 1982).
Retrospective Research
Taylor (1978) studied life events and herpes recurrences in a
sample of 60 female students, of whom 36 had genital herpes and 24
served as controls. All were administered the Life Experiences Survey
and a Genital Herpes Questionnaire. Women with genital herpes did not
report a higher frequency of stressful events over the prior year than
women without herpes. Within the herpes groups, however, women with a
high rate of recurrence (four or more in previous year) were compared
to those with a low recurrence rate (three or less). The high
recurrers reported having significantly more negative events than the
low recurrers. Due to the retrospective nature of the study, it is
impossible to know whether high levels of negative events were a cause
or effect of herpes reactivations. In the same study, 70% of the
women indicated that having genital herpes caused emotional stress.
Additionally, 92% of the herpes patients recalled experiencing stress
the week prior to a recent occurrence, while only 42% of normal
controls recalled stress at a comparable period. The potential
retrospective bias is apparent. In addition to patients' expectations
regarding stress and illness, they had a dramatic event (herpes
episode) to tie their recollections to while control subjects may not
have had a temporally contiguous event to aid in their recall of
stresses occurring at that time. These findings are supportive of a
stress/recurrence relationship, but are limited by retrospective bias,


27
the lack of objective documentation of the illness measure, and the
restricted nature of the sample (female students).
Watson (1983) extended Taylor's (1978) findings in an
investigation of 51 male and female volunteers who had at least one
recurrence of genital herpes within the 7 months prior to
participation. A significant positive relationship was found between
undesirable life events in the previous year as reported on the Life
Experiences Survey and genital herpes recurrences in the previous 6
months. Recurrences in the prior 6 months were also significantly
related to undesirable life events reported for the 12- to 6-month
period prior to the study, suggesting a long term effect of major life
events on herpes recurrences.
Watson (1983) is the only investigator to consider potential
moderators of life stress and herpes recurrences. He included
measures of perceived locus of control and social support in his
research design. Using the Rotter Internal/External Control Scale
(Rotter, 1966), Watson (1983) demonstrated an interaction between
locus of control and stress in which subjects who perceive internal
control experienced fewer recurrences under high stress conditions
than those with an external locus of control. This interaction effect
suggests a moderating influence of locus of control on the negative
effects of stress. Social support as measured by total helpfulness on
the Social Relationship Scale did not appear to have a similar
moderating effect on stress. However, a more direct effect was
found. Subjects reporting high levels of helpfulness had
significantly fewer recurrences than those reporting low helpfulness


28
(Watson, 1983). This investigation emphasizes the importance of
investigating potential moderating factors. It is limited by reliance
on retrospective reporting and the nonrepresentative nature of the
sample, primarily high recurrers recruited from a self-help support
group.
Prospective Research
A series of prospective investigations have been undertaken to
examine the relationship of psychosocial variables and recurrences of
oral HSV (Friedman, Katcher, & Brightman, 1977; Katcher, Brightman,
Luborsky, & Ship, 1973; Luborsky, Mintz, Brightman, & Katcher,
1976). Independent measures included scores on the Life Change Index
reported retrospectively over the 2 years prior to the study, the
Cornell Medical Index, John Hopkins Symptoms Index, Clyde Mood Scale,
a Social Assets Scale, illness history, and blood analysis. Dependent
variables included illness records, infirmary visits, and, when
possible, documentation of HSV recurrences.
A year long investigation of 67 paid volunteers, 14 of whom had
significant antibodies to HSV, found previous history and blood serum
antibody presence most predictive of oral herpes recurrence, with the
unhappy factor of the mood scale also contributing significantly
(Katcher et al., 1973). When only the psychological variables were
considered, the unhappy factor accounted for 16% of the variance. The
relationship between predictors was different for herpes and systemic
illness, suggesting that recurrences do not reflect an illness
disposition. A significant negative relationship of social assets and


29
herpes episodes implies a possible moderating effect of social
assets. The 2-year retrospective life events measure was of minimal
value in all analyses and was disregarded in the subsequent
investigations (Friedman etal., 1977; Luborsky etal., 1976). While
life events were not an important variable in this study, this may be
explained on the basis of methodology. The number of subjects who
were seropositive for herpes was small. After attrition, only 33
subjects completed a full year of participation. The stepwise
multiple regression procedures were based on 19 episodes of herpes,
only 10 of which were documented by examination and viral culture.
The life events measure may have been insensitive due to memory
decline (Monroe, 1982a), the failure to consider negative events
separately (Dohrenwend & Dohrenwend, 1974; Johnson & Sarason, 1984),
or a lack of association between temporally remote events and
recurrences. However, the investigators anecdotally report an
increase of herpes recurrences during exam time for students (Katcher
et al., 1973).
A 3-month daily investigation of 20 herpes seropositive subjects
failed to demonstrate any systematic relationships between mood and
herpes reactivations (Luborsky et al., 1976). However, almost all
subjects reported stresses prior to episodes when responding
retrospectively. The authors suggest that by having moods reported in
the morning, the emotional impact of the day was missed. But it may
also be the case that moods are not necessarily equivalent to stress.
The researchers then followed a larger group of participants for
3 years (Friedman et al., 1977), during which 51 of 149 had at least


30
one HSV episode. Using herpes incidence as the dependent variable in
a multiple regression analysis, the social and psychological variables
were found to be of least predictive value (3-5% of explained
variance), biological factors (disease history and illnesses,
especially upper respiratory infection) being of greatest value
(80-90% of explained variance). Unfortunately, these investigators
did not include a stress measure. They replicated the earlier finding
that predictors were specific for herpes and unrelated to predictors
for upper respiratory infection (Friedman et al., 1977; Katcher et
al., 1973).
Daily samples of vaginal secretions from five women, three of
whom were positive for genital herpes were collected for a month by
investigators of asymptomatic viral shedding (Adam, Dressman, Kaufman,
& Melnick, 1980). Herpes simplex virus was isolated at least once
from each of the women with a history of herpes. Multiple positive
results by immunoperoxidase staining of cervical-vaginal smears were
found for all three. No statistical analyses were performed, but
graphical representation revealed temporal clusters of positive
results for each woman. These were anecdotally related to
menstruation, emotional stress, exam periods, and problems with
children. Unfortunately, the stress data were not collected in a
systematic manner making findings difficult to interpret (Adams et
al., 1980).
The data from a group of 58 psychiatric patients with primary
episodes of genital herpes who were followed prospectively for 28
weeks by Goldmeier and Johnson (1982) provide some support for a link


31
between psychosocial factors and recurrence. Subjects completed a 60-
item screening measure of psychiatric symptoms, the General Health
Questionnaire, on which they indicated which feelings and behavior
(symptoms) had been experienced over the past 4 weeks. They were then
requested to return to the clinic for viral cultures in the event of a
recurrence over the next 28 weeks. Subjects who did not return were
mailed a follow-up form requesting recurrence information. Of the 29
subjects reporting a recurrence, 7 returned for culture confirmation
of herpes. General Health Questionnaire Scores of recurrers were
compared to the nonrecurrers. The nonrecurrers had significantly
lower scores than the recurrers. Furthermore, the 29 subjects above
the cut-off score of 11 which indicates potential or overt psychiatric
problems had a significantly higher recurrence rate as determined by
actuarial recurrence-free curves. Investigators infer that anxiety or
obsessionality may increase production of adrenergic substances and
contribute to herpes reactivation. This hypothesis is consistent with
literature implicating the autonomic nervous system in decreases in
immune responsivity (Hall & Goldstein, 1981; Rogers, Dubey, & Reich,
1979).
Summary
To summarize, despite the widespread popular notion that stress
and recurrences of herpes are linked, consistent empirical support for
such a relationship is lacking. A few anecdotal reports suggest a
connection between psychological distress and recurrences (Blank &
Brody, 1950; Schneck, 1947) and hypnotic suggestion is reported to


32
influence recurrences (Janicki, 1971; Ullman, 1947). Strong placebo
effects have been apparent in controlled investigations (Hamilton,
1980; Kern, 1979). Survey research demonstrates that patients with
recurrent herpes believe there is a connection with emotional stress
(Bierman, 1982; "Help survey," 1931). In retrospective investigations
of patients with genital herpes, high recurrence rates have been
associated with high major life stress (Taylor, 1978) and, in one
study, negative life stress appears to have exerted a long term effect
on recurrence rate (Watson, 1983). Unfortunately, anecdotal reports,
survey research, and retrospective studies are all subject to
distortion.
In one prospective investigation of oral herpes recurrences, a
negative mood factor contributed significantly to recurrence rate, but
a life events measure proved to be unimportant (Katcher et al.,
1973). Moods failed to hold up as important in subsequent prospective
studies in which life stress was unfortunately not measured (Freidman
et al., 1977; Luborsky et al., 1976). Anecdotally related stress
episodes were graphically associated with laboratory evidence of viral
shedding for three women (Adams et al., 1980). And finally, patients
demonstrating a strong potential for psychiatric problems on a
screening device subsequently reported more recurrences than those
falling below the critical cut-off score, suggesting a connection
between anxiety and recurrences (Goldmeier & Johnson, 1982). So far,
empirical support for a stress/herpes association is suggestive rather
than strong. A longitudinal prospective study with regular,


33
systematic collection of stress data and laboratory documentation of
recurrences would contribute valuable information.
In the following section, an attempt will be made to argue for
the significance of such a study in the broader scheme of
health/illness research. A reasonable proposal will be made of the
sequence of environmental, psychological, and physiological events
which could take place if stress is, indeed, a significant factor in
precipitating herpes recurrences.
Theoretical Framework
Perception of Stressful Events
Between an environmental event and a recurrence of genital herpes
many levels of complex interactions are possible. Figure 1
illustrates in simplified form the hypothesized process by which
environmental events could result in recurrences of genital herpes.
First, there must be a stimulus, the occurrence of a major or minor
life event. For it to be reported as a stressor or a hassle, it must
be perceived as such. Moderating influences may affect an
individual's perception of events in either negative or positive
ways. A person who craves excitement would probably view the event of
a flat tire enroute to work differently than a counterpart who thrives
on routines and predictability. A person who feels cared for and
supported by family and friends will probably be less threatened by
criticism on the job than a more solitary individual. The point is
that the same event could be perceived as a stressor by one
individual, but not another.


Simplified model of some of the hypothetical levels through which environmental events may
affect biological events and ultimately result in a herpes recurrence.




36
Once an event is perceived as a stressor, it may initiate a
variety of physiological responses as the individual struggles to
adapt to the new situation. Just as moderator variables may influence
a person's perception of stressors, they may influence biological
reactions as well. Internal adaptation responses may occur in the
nervous, endocrine, and immune systems. Awareness of internal
responses may, in turn, affect perception in a feedback loop in a
manner to exacerbate or diminish the experience of threat. The
resultant fluctuation may upset the delicate balance necessary for
herpes simplex virus to remain in its quiescent existence. The virus
could then reactivate and result in an episode of recurrent herpes.
In the following section, potential mechanisms by which perceived
stress could result in a herpes recurrence will be considered.
Proposed Physiological Mechanisms
The precise triggering mechanism which leads to reactivation of
herpes simplex virus is elusive. While residing in the ganglion, the
virus neither destroys the latently infected cells nor is eliminated
by host defense mechanisms or prolonged antiviral chemotherapy.
Something occurs at the molecular level to stimulate it to migrate to
the skin surface and begin viral replication (Bierman, 1983; Stevens,
1975). To explain this phenomenon, two major hypotheses have been
proposed. In one hypothesis, the dynamic state hypothesis, a small
number of cells constantly replicate virus. Specific immune defense
responses involving antibody, lymphocytes, and interferon continually
work to localize and prevent the clinical manifestations of recurrent


37
infection (Stevens, 1975). This theory, also called the "skin trigger
theory," proposes that recurrences result when local defenses are
temporarily suppressed, or when slight physiological changes (other
than immune mechanisms) alter the balance in favor of the virus (Hill
& Blyth, 1976).
The static state hypothesis or "ganglion trigger" theory proposes
that the virus remains inactive in the ganglia until nervous system
stimulation provokes it to migrate toward the skin surface. In this
theory, the viral genome remains in an unproductive, quiescent state
(Stevens, 1975). At present, the weight of evidence favors the static
state hypothesis. Investigators that once believed in the dynamic
hypothesis (Hill but inactive in the ganglion. It is repressed and expresses few if
any viral antigens and, therefore, provokes no significant immune
response unless infection is reactivated (Blyth & Hill, 1984; Hill,
Blyth, Harbour, Berrie, & Tullo, 1983). The precise mechanism of
derepression allowing viral expression remains a mystery.
There is some indication that immune system status influences an
individual's vulnerability to recurrent herpes. This is particularly
true for cell-mediated; immune responses as opposed to humoral
immunity. Once infected with herpes simplex virus, individuals
develop neutralizing antibodies (humoral immunity) within 1 to 4 weeks
of primary infection (Nahmias & Roizman, 1973c). Unfortunately, the
antibodies are not effective in preventing reinfection or recurrent
infection in either healthy or immunocompromised individuals (Nahmias
& Roizman, 1973a; Rand, 1982; Rand, Rasmussen, Pollard, Arvin, &


38
Merigan, 1977). The herpes viruses, being capable of cell to cell
spread, are well adapted to evade antibodies. It has been
demonstrated that the titers of neutralizing antibody only
occasionally fluctuate before, during, or after recurrences (Nahmias &
Roizman, 1973b).
Fluctuations in cell-mediated immune responsivity, on the other
hand, appear to be related to recurrent herpes infections. Evidence
for this relationship comes from a variety of sources. A defect in
cellular immunity is proposed to account for the severity of herpes
infections in the newborn and the irnmunologically impaired host
(Nahmias & Roizman, 1973c). A quicker, more intense in vitro
lymphocyte transformation response was related to faster healing in
primary infection in one investigation. Thirteen of the patients were
followed for a period which included one or more recurrence. The
herpes simplex virus stimulation index was declining when recurrences
took place. Because the stimulation index was declining even among
patients who did not experience recurrence, authors suggest this
decline may be a necessary but insufficient correlate of recurrence
(Corey, Reeves, & Holmes, 1978). Similar results were obtained in an
investigation comparing the lymphocyte responsivity of patients with
frequently recurring oral herpes to a control group of seropositive
blood bank donors not reporting frequent herpes recurrences (Kirchner,
Schwenteck, Northoff, & Schopf, 1978). Responsivity of the two groups
was similar during infection, but during the disease-free interval
responsivity of the patient group was significantly lower. The
severity of infection in immunocompromised patients has been related


39
specifically to decreased responsivity of lymphocytes as well (Meyers,
Flournoy, & Thomas, 1980; Rand et al., 1977).
Another parameter of cellular immunity, the specific cytotoxicity
of peripheral blood mononuclear leukocytes (PBML) to cells infected
with herpes was found to fall to low levels during the week prior to
recurrent HSY in a group of 14 healthy seropositive adults, showing
enhanced reactivity during acute recurrent infection (Rola-Plezczynski
& Lieu, 1984). T-cells and the percentage of natural cytotoxic
lymphocytes were also found to decrease prior to recurrent HSV
episodes and to increase during recurrences. Fluctuations in
interferon, another component of cellular immunity, have likewise been
implicated in recurrences (Corey & Holmes, 1983; Rand et al., 1977)
and administration of interferon has significantly reduced the risk of
reactivation of patients undergoing surgery on the trigeminal ganglion
(Pazin et al., 1979).
To summarize, it appears that humoral immunity does not play a
major part in the manifestation or recovery from recurrent HSV
infection. This does not rule out the possibility that antibody may
contribute in combination with complement or as a mediator of
lymphocyte cytotoxicity. On the other hand, a relationship to HSV
infection status has been demonstrated for a number of cell-mediated
immune parameters including quantity and responsivity of T-
lymphocytes, interferon release, and natural lymphocyte cytotoxicity
specific for HSV.
A deficiency in immunologic competence, particularly impairment
of cell-mediated responsivity, could be a link between psychosocial


40
stress and recurrent herpes. Alterations as a result of stress have
been demonstrated in parameters of cellular immunity such as decreases
in lymphocyte responsiveness and natural killer cell activity
(Bartrop, Luckhurst, Lazarus, Kiloh, & Penny, 1977; Locke, 1982;
Kiecolt-Glaser etal., 1984). Changes in cel 1-mediated immune
functions are in turn related to herpes recurrences. An individual's
response to stress, therefore, could cause immunosuppression which
would allow viral replication. Physiological evidence for such a
connection will be reviewed as well as evidence for a more direct path
of influence which does not necessarily involve immune status.
There is consistent evidence that immunosuppression has an effect
on recurrences of herpes simplex virus. Seropositive transplant
patients whose immune responses are pharmacologically suppressed to
prevent organ rejection experience herpes lesions that are
inordinately severe, extensive, and long lasting. This occurs
sometimes, but not always, with increased incidence over a presurgery
recurrence base-rate (Blyth & Hill, 1984). For example, the
recurrence rate 5 to 15 weeks after bone marrow transplant is 60-80%
(Meyers et al., 1980). In the first 3 months following cardiac
transplantation, 83% of seropositive patients experienced herpes
recurrences of prolonged duration averaging over 2 months in duration
(Rand et al., 1977). Fifty percent of the seropositive renal
transplant paients developed herpes lesions of 2 to 16 weeks duration
within a 4-month follow-up period (Pass et al., 1979).
One of the early investigations to indirectly implicate stress as
a factor in the reactivation of HSV was done on rabbits. Laibson and


41
Kibrick (1966) experimentally induced recurrent ocular herpes in
latently infected rabbits by intramuscular administration of
epinephrine, a hormone well known to be released in excess during
periods of stress. Investigators demonstrated significant viral
isolation and infection in experimentally inoculated animals relative
to control animals. One could postulate an autonomic nervous system
link between stress and recurrence based on this evidence.
Reactivation of herpes in animals and humans has also been
accomplished by direct neural stimulation. Low levels of electric
current delivered via electrodes implanted over the trigeminal
ganglion of latently infected rabbits consistently induce viral
shedding and herpes recurrences, depending on the interval between
stimuli (Green, Rosborough, & Dunkel, 1981). In humans, trigeminal
surgery appears to exert a direct triggering effect. Between 83-94%
of seropositive individuals who undergo trigeminal surgery experience
recurrence and/or viral shedding within a week. These high incidence
recurrences are unremarkable in duration and severity and are
restricted to the area supplied by the stimulated or sectioned portion
of the trigeminal sensory ganglion (Carton & Kilbourne, 1952; Pazin et
al., 1979). Excessive; stimulation to peripheral tissue which receives
sensory nerves from the infected ganglion induces reactivation in
animals (Hill et al., 1983). Such stimulation could be analogous to
reactivation by sun, wind, or physical injury in naturally occurring
herpes episodes. One could extend the concept of reactivation by
direct neural stimulation to include excessive neural firing as might
result from psychological as well as physical stimulation. Thus, just


42
as hypnosis or expectancy (placebo effects) may act directly to alter
recurrence rates via a postulated central nervous system mechanism,
anxiety and stress may also act directly to increase recurrence rates
by causing excessive neural firing or by lowering the threshold for
neural firing.
It should be noted here that there appears to be a subtle
difference in how direct neural stimulation and immunosuppression
influence herpes recurrences. Direct neural stimulation results in
more immediate and higher rates of recurrence. These directly induced
recurrences are of average extent and duration. The recurrence rate
in immunosuppressed individuals is not necessarily higher, but the
recurrences are more extensive and long lasting. Of course, any
surgical treatment, whether immunosuppression is involved or not,
involves both physical and psychological trauma, providing a
confounded picture. Presumably, patients undergoing trigeminal or
transplant surgery experience grossly similar psychological
responses. A comparison of herpes simplex virus activation in these
two groups of patients suggests that direct neural stimulation is
associated with reliable reactivation of the virus, while
immunosuppression is more frequently associated with prolonged
severity and duration of recurrent herpes episodes.
One could conjecture more than one route by which stress could
influence the reactivity of HSY. Stress could directly stimulate the
latent virus via the central nervous system by causing increased
neural firing or a lower threshold for neural firing. Stress could
also be indirectly involved, by causing a deficiency in immune


43
surveillance. It is also possible that both are necessary to provoke
the clinical manifestation of herpes symptomatology, a trigger to
reactivate the virus and an immunocompromised environment in which it
may flourish. At the present time, the exact mechanisms that control
the latency and reactivation of herpes simplex virus in humans are
unknown. Immune response could play a vital role, but the
physiological state of the neuron could also be a key factor (Blyth &
Hill, 1984).
Because of its compatibility with existing evidence, the
supposition that both neural stimulation and immunosuppression are
necessary for herpes reactivation is intuitively appealing. However,
examinations of the validity of this and the other hypotheses present
serious difficulties. While it is known that some triggers are more
effective than others (e.g., trigeminal surgery more frequently
results in reactivation of HSV than sun exposure), the exact nature of
neural stimulation necessary to provoke a recurrence is unknown and,
therefore, difficult to measure or manipulate. Likewise, differences
in immunocompetency are difficult to define and even more difficult to
manipulate, particularly in humans, due to the ethical issues
involved. The preferred course at present is to continue research at
all levels of organization from the precise cellular/biochemical
measurements obtainable in the lab to the more global psychosocial
measurements obtainable through systematic naturalistic observations
of individuals over extended time periods. An integrated effort by
professionals with expertise in immunology, endocrinology, neurology,
pathology, and the behavioral sciences will be necessary to discern


44
meaningful patterns from divergent multiple measures. Ideally,
converging evidence from all of these disciplines will bring us closer
to the answers.
Present Investigation
The present study was undertaken to extend our knowledge of the
relationship between stress and health status. It was designed to
address methodological shortcomings of existing investigations.
Genital herpes was chosen as a disease model because of its recurrent
nature, the theoretical basis for a physiological mechanism linking
stress and recurrences, the compelling need for empirical information
regarding stress and recurrences, and the availability of an objective
laboratory diagnostic measure for disease confirmation. Prospective
in nature, the study design called for an evaluation of major and
minor life events upon enrollment and on a monthly basis throughout
the 6-month period of investigation. Documentation of herpes lesions
occurred within 24-36 hours of each recurrent episode. Thus, both
prospective and concurrent information regarding stress and
recurrences was collected. Additionally, three variables which have
shown promise in prior' research as moderators of life stress were
assessed. The use of multiple regression techniques enabled the
clarification of the relative importance of social support, locus of
control, and arousal seeking tendency in influencing recurrence rates
in the face of stress. Thus the possibility of identifying subgroups
of individuals who were thought to be particularly vulnerable was
explored. In the present state of the art of psychobiological


45
research in which the single-cause, single-effect model is
acknowledged as simplistic, we must depend upon converging evidence
from the different levels of many disciplines to obtain a reasonable
picture of the many internal and external events and dynamic
interactions that ultimately result in a disease outcome (Schwartz,
1982).
This particular investigation involves the second and fifth
levels as illustrated in Figure 1: the perception of events as
stressful or not and the occurrence of viral reactivation as
manifested in recurrent herpes episodes. No attempt was made to
objectively monitor environmental events in the lives of subjects by
trained neutral observers. Nor have attempts been made to measure
neural, endocrine, or immune parameters. The independent variables in
this study consist of self-report measures of major and minor
stressors and social and personality variables which hypothetically
moderate the perception and influence of the stressor variables. The
dependent variable measured was a manifestation of genital herpes
symptomatology, the isolation of virus from a lesion. While this
particular investigation has not attempted to elucidate the complex
web of physiological events between stressor and symptomatology, it
has endeavored to shed light on the psychological conditions under
which a relationship between stress and illness exists.
Hypotheses
Several specific hypotheses were investigated in the present
investigation. As stated earlier, the present investigation was


46
designed to explore the relationship between stressful events and
recurrences using both prospectively and concurrently collected
information. The first hypothesis, using a prospective approach,
concerns the strength of the relationship between an individual's
experience of major life events and subsequent experience of herpes
recurrences. Specifically, the first hypothesis states:
1. The occurrence of undesirable life events prior to entry into
the study will be positively related to herpes recurrences during the
6-month period of investigation.
The next two hypotheses concern the concurrent experience of both
major and minor life events and herpes recurrences. Specifically,
2. The occurrence of undesirable life events during the 6-month
period of investigation will be positively related to herpes
recurrences over the same period of time.
3. The occurrence of monthly hassles will be positively related
to herpes recurrences during the period of investigation.
Additionally, it was hypothesized that three moderator variables,
social support, locus of control, and arousal seeking tendency, would
interact with each other in such a way that an individual's
vulnerability to herpes recurrences under high stress conditions would
be predictable on the basis of scores on the moderator variables.
That is,
4. The following variables which have previously been shown to
moderate the effects of life stress will be related to HSV recurrences
in the following ways: (a) Low levels of social support, a perception
of events as externally controlled, and a tendency to avoid arousing


47
situations will be associated with more frequent HSV recurrences, (b)
The antithesis of the above will serve to protect individuals from
frequent HSV recurrences. That is, individuals with a high level of
social support, with an internal perception of control, who tend to be
arousal seeking will have fewer HSV recurrences, (c) The probable
frequency of recurrence for a given individual will be predictable on
the basis of scores on the moderator variables.


CHAPTER TWO
METHODS
Research Design
The longitudinal design of this study allowed for the
investigation of prospective and concurrent information regarding the
occurrence of major and minor stressful events and episodes of
recurrent genital herpes. Data concerning past stressful events
collected from initial interviews and questionnaires provided
information to relate to ensuing recurrences of genital herpes
prospectively documented over the 6-month period of the study.
Concurrent information on major and minor stressful events was
collected monthly during a subject's 6 months of participation to
relate to documented recurrences for the same period of time. This
design provided the opportunity to avoid distortion due to
retrospective bias in an investigation of the relationship of major
and minor stressful events and herpes recurrences.
The assessment of; subjects' sense of perceived control, tendency
toward arousal seeking, and social support provided an opportunity to
evaluate the hypothesized stress-moderating effects of these
variables. In addition to studying the effects of the presumed
moderator variables separately, multivariate statistical techniques
made it possible to evaluate the manner in which these variables
related in an additive or interactive manner to possibly enhance or
48


49
diminish the resistance of individuals to recurrent episodes of
genital herpes in the face of life stress.
Sample Selection
Subjects were recruited over a 16-month period (March, 1983-July,
1984) through advertisements and referrals. Advertisements were
placed in both the university and local newspaper and on bulletin
boards on campus and in the J. Hillis Miller Health Center. Sources
of referral included the University of Florida Student Health Center,
the clinics of Shands Teaching Hospital, and participating
volunteers. Subjects were telephone-screened regarding their
suitability, availability, and willingness to make a 6-month
commitment to the sudy. The research was described as an effort to
understand more about the psychological factors involved in the
recurrence of genital herpes in order to improve the counseling
techniques of health care providers.
After obtaining informed consent, 10 cc of blood was drawn from
each participant. Those with a positive antibody titer to herpes were
enrolled in the study. Those subjects with a negative titer were not
enrolled unless they had a positive culture from a genital site to
herpes simplex virus.
Each of the 122 subjects who completed the full 6 months of
filling out questionnaires and having genital herpes recurrences
cultured received 100 dollars at the conclusion of the study for
his/her participation.


50
Measures
Stress Measures
The occurrence of major and minor events in the lives of subjects
was assessed by self-report questionnaires. Subjects completed the
questionnaires upon enrollment and monthly during the study.
Life Experiences Survey (LES). This questionnaire was used to
evaluate the experience of subjects with major life events. The LES
is comprised of 57 life events, 3 of which are fill-ins and the last
10 of which are specifically relevant to a college population
(Sarason, Johnson, & Siegel, 1978). Subjects indicate the frequency
of occurrence of each of the events over a specified time period.
They also rate the impact of each event on a scale from -3 (extremely
negative) to +3 (extremely positive), a zero rating designating a
neutral impact. Means and standard deviations are available for the
47- and 57-item versions based on 345 college students. Since many
subjects were college students, the longer inventory was used.
However, group statistical analyses were based upon the 47 items which
all subjects completed.
Desirable and undesirable events are assessed separately on the
LES, an advantageous feature for this investigation. Negative events
have been found to be more predictive of illness in general (Hotaling
et al., 1978; Sheehan et al., 1979) and genital herpes in particular
(Taylor, 1978; Watson, 1983). This is consistent with the literature
on adjustment (Mueller et al., 1977; Vinokur & Selzer, 1975).
Although impact (subject weighted) scores were available, the measure
of interest in this investigation was the frequency of undesirable


51
events. Several studies providing data on both simple frequency
scores and weighted scores have shown them to be equally predictive
where group data are concerned (Monroe, 1982b; Skinner & Lei, 1980;
Zimmerman, 1983). Developers of the scale have provided information
regarding significant positive correlations with several different
measures of anxiety, noncomformity, discomfort, depression, locus of
control, and a nonsignificant correlation with the Marlowe Crowne
Social Desirability Scale to demonstrate validity. Additionally, the
scale has been found to discriminate between normal controls and
counseling clients (Sarason et al., 1978). This scale has been used
in two retrospective investigations of subjects with recurrent
herpes. In one, the scale discriminated between subjects experiencing
high and low recurrences (Taylor, 1978). In the other, significant
positive correlations were found between the frequency and impact of
undesirable life events of the past year and the frequency and
duration of herpes recurrences for the past 6 months (Watson, 1983).
Hassles Scale (HS). The HS focuses on the minor events of the
past month, the irritating, frustrating, and distressing demands
characterizing daily transactions with the environment (Kanner et al.,
1981). The 117 items which include such things as care for pet,
having to wait, inconsiderate smokers, too many meetings, problems
with children, and preparing meals are checked if perceived as a
hassle and rated from 1 to 3 according to the severity of the
hassle. A frequency score is obtained by counting the number of
hassles checked. An impact score is obtained by summing the hassle
ratings. Norms are based on 100 middle class subjects ranging in age


52
from 45 to 64 who completed the scale for 9 consecutive months.
Test-retest reliability over that period is reported as .79 for
frequency scores, .43 for impact scores. Significant positive
correlations have been found with a negative affect scale, a life
events measure, a symptom check list (Kanner et al., 1981), and a
general health measure (DeLongis et al., 1982).
Moderator Variables
Upon enrollment in and termination from the research 6 months
later, subjects completed inventories to assess enduring
characteristics proposed to moderate the effects of stress on
illness. These included measures of social support, of perception of
internal or external controllability of events, and of arousal seeking
tendency.
Locus of Control (LC). The LC is a 29-item forced-choice
inventory including six filler items which assesses the degree to
which individuals view environmental events as being under their
personal control (Rotter, 1966). Subjects scoring low on the LC
(internals) tend to perceive events as being controllable by their own
actions, while those scoring high (externals) tend to view events as
t
being influenced by factors other than themselves. It is a measure of
generalized expectancy, or belief, rather than preference. The
measure is widely used and extensive data are available regarding
internal consistency, test-retest reliability, and discriminant
validity. Correlations with social desirability and intelligence are
low. Means and standard deviations on the scale are provided for


53
numerous groups, including large groups of university students
(Rotter, 1966). A number of studies investigating psychological
adjustment and illness as related to life stress provide support for
the role of perceived control as a moderator of life stress. That is,
an external LC score is associated with instability and illness under
conditions of increased life stress (Johnson & Sarason, 1978; Manuck
et al., 1975; Toves et al., 1981).
Arousal Seeking (AS). Arousal seeking is defined as the tendency
of individuals to engage in or avoid situations which increase
stimulation to a personal optimal level. The AS scale was developed
specifically to assess individual differences in this trait (Mehrabian
& Russell, 1974). Subjects are instructed to indicate their degree of
agreement or disagreement (on a 9-point scale) to each of 40 items.
Means and standard deviations are reported for two large samples of
college students. Adequate internal consistency and reliability have
been demonstrated. The AS scale correlated positively with a measure
of extroversion and negatively with measures of anxiety.
(Correlations with anxiety are significant, but weak; it is not simply
an anxiety measure.) Low correlations have been found with social
desirability and succorance (Mehrabian & Russell, 1974). In an
investigation of the relationship between life stress and measures of
anxiety, depression, and hostility, a significant relationship was
found only for subjects low on the arousal seeking dimension.
Negative life change in subjects who were low in arousal seeking
resulted in greater anxiety and hostility. This suggests that the


54
dimension plays a moderating role on life stress (Johnson, Sarason, &
Siegel, 1979).
Social Support Questionnaire (SSQ). Social support may be
defined as the degree to which individuals have access to social
resources or relationships upon which they can rely, particularly in
times of need. A person's satisfaction with the quality of such
resources is an important dimension in addition to the number of
persons in a support system. The SSQ measures the size and
satisfaction rating of an individual's social network (Sarason,
Levine, Basham, & Sarason, 1981). It consists of 27 two-part items,
one part to indicate the number of support people available under
particular circumstances, the other to indicate the respondent's
satisfaction rating for available support. Norms exist for a sample
of 602 university students. Internal consistency and test-retest
reliability are high. Correlations with scales of maladjustment show
an inverse relationship, suggesting that people who have fewer and
less satisfying social supports are more likely to be anxious, more
labile emotionally, and more pessimistic than others. Subjects low on
the SSQ tend to have an external locus of control and be relatively
low in self-esteem. '
There have been several studies which provide evidence for social
support as a moderator of life stress (de Araujo et al., 1973;
Hotaling et al., 1978; Nuckolls et al., 1972; Schaefer et al.,
1981). Those low in social support evidenced more maladjustment and
ill health under circumstances of high life stress. Therefore, high
levels of social support may play a stress-buffering role. In the


55
present investigation the SSQ was modified to include a pair of
questions regarding to whom a subject had disclosed he/she had herpes
and level of satisfaction with support received.
Physiological Measures
Screening procedure: HSV antibody. Since enrollment in the
study was limited to those who were herpes simplex virus (HSV)
seropositive, subjects were screened in order to ensure that they had
had prior exposure to HSV. This work was performed in a virology
laboratory and was done by a microtiter complement fixation (CF)
method previously described (Rand, Kramer, & Johnson, 1982). No
attempt was made to discriminate between antibodies to herpes simplex
virus-1 (HSV-1) and herpes simplex virus-2 (HSV-2) because (a) sera
that is seropositive by CF is seropositive for both HSV-1 and HSV-2 by
an accepted discrimination technique (ELISA); and (b) although HSV-2
usually causes genital herpes, HSV-1 is also a frequent cause.
Dependent variable: Virus isolation and transport. All subjects
were instructed to call the Clinical Research Center (CRC) as soon as
they were aware of a recurrent genital herpes lesion. An appointment
was made for a culture to be obtained within 24-36 hours by trained
i
personnel in the CRC. The CRC personnel were available on a 24-hour
daily basis.
Cultures for HSV were obtained by swabbing the base of unroofed
vesicular or ulcerated lesions with a cotton swab. Swabs were
immersed in transport media [minimal essential medium (MEM)
supplemented with 2 mm glutamine, 5000 u/1 penicillin, 50 ug/ml


56
streptomycin, 100 ug/ml gentamicin, 2 ug/ml amphotericin B and 1%
gelatin] and held at 4C until transported to the laboratory; previous
work in the laboratory (unpublished) and elsewhere has shown this
method does not result in any significant decrease in recovery of
HSV-2 (Bettoli, Brewer, Oxtoby, Azidi, & Guinan, 1982; Yaeger, Morris,
& Prober, 1977). Cultures were then either frozen at -70C or
inoculated immediately into primary cultures of human foreskin
fibroblasts (Rand et al., 1982). All isolates were identified by
typical cytopathologic effect.
Because subjects would not always be able to come to the CRC in
time, they were given Virocult swabs (Medical Wire and Equipment Co.,
Cleveland, OH) upon enrollment and instructed in the procedure for
obtaining self-cultures. Pilot work indicated that patients self-
obtained cultures were positive as frequently as staff-obtained
cultures. Herpes simplex virus has been shown to be stable for at
least 48 hours when refrigerated under these conditions (Medical Wire
and Equipment Co., product information).
Procedure
Initial Contact ;
Subjects were seen singly at the CRC by the investigator or a
trained research assistant. Confidentiality was ensured, study
procedures were explained in detail, questions answered, and informed
consent obtained. Blood was drawn for screening and subjects were
instructed to return to the CRC to have herpes lesions cultured every
time they had a recurrence during the 6-month period of the study.


57
They were provided with a Virocult swab and trained to obtain a self
culture in the event that it was impossible to get to the CRC within
24-36 hours. A culture could be safely kept in the refrigerator for 2
to 3 days until transport to the clinic was possible. A sample copy
of the LES was provided with a calendar on which to record major life
events. This eased the filling out of questionnaires each month and
helped to avoid overlap or duplication of major life events from month
to month. Subjects were then given a packet of initial questionnaires
and instructed in the procedure for completion. The packet contained
the LES (for the past 6-month and 1-year periods), the HS, LC, AS, and
SSQ questionnaires. An appointment was made for the second contact in
1 to 2 weeks time when the blood screening results would be available.
Second Contact
Subjects returned with the completed questionnaires of the
initial packet. Their questions were answered, and they were given a
second opportunity to sign the consent form to signify continuing
commitment to the investigation at a time when blood screening results
were known and they were thoroughly familiar with all procedures and
questionnaires. Subjects were not considered fully enrolled until
I
their second signature was obtained. Completed questionnaires were
reviewed for proper completion and a structured interview was
conducted to obtain demographic information and the subjects' history
of herpes. (Please see Appendix A for a copy of the structured
interview.)


58
Continuing Contact
Continuing contact was by mail, phone, and clinic visits. In the
last week of each month, subjects received by mail a packet containing
precoded questionnaires (LES and HS) to be completed on the final day
of the month with respect to events of the month; self-addressed,
stamped envelopes for their return; a newsletter; and educational
materials regarding genital herpes to maintain motivation. Subjects
were requested to date the major events reported on the LES to avoid
overlap with preceding and following months. Phone contact was made
to subjects to clarify material on the questionnaires, or to prod for
their return as necessary. Subjects initiated phone contact at any
time to ask questions about the study, or to obtain information about
herpes. Clinic visits occurred at any time during the 6 months of
participation for the purpose of obtaining culture material from
recurrent lesions. Subjects received counseling for disease
management by the CRC staff or the investigators if appropriate during
clinic visits.
Final Contact
A debriefing interview was conducted with each subject as soon as
the questionnaires for the sixth month were completed. Payment was
made to subjects following this final interview.


CHAPTER THREE
RESULTS
Sample Characters sties
The original sample consisted of 153 individuals. Five were
excluded because they did not have HSV antibodies in the initial
screening and thus were not subject to recurrent HSV. Attrition
occurred for the following reasons: (a) moved, j^=8; (b) too busy,
n_=4; (c) lost interest, jr=2; (d) received experimental vaccine, n_=l;
(e) believed participation worsened condition, n_=l; (f) visual
handicap, _n_=l; and (g) lost to follow-up, n_=8. Considering only the
seropositive subjects, this represents a loss of 26, or 18%, of the
eligible subjects.
The final sample consisted of 122 subjects, 83 of whom were
female (68%) and 39 of whom were male (32%). Average age was 27.3
(SD=6.1) with a range of 18 to 55. Educational level was high, with
the average number of years of education being 14.3 (range=9-18).
Sixty-three (52%) of the participants were currently in college or
graduate school. The majority of the participants (94; 78%) were
single, although 83 (68%) reported having a regular sex partner.
Subjects reported having herpes an average of 3 years, 11 months
(range of 1 month to 18 years). For the year prior to the study,
46.7% reported four or more recurrences, 35.8% reported one to three
recurrences, and 17.5% reported none.
59


60
Undesirable Life Events
The measures of major and minor life events in this study each
produced two scores of potential relevance to the present
investigation, a simple frequency of negative events score and an
impact score reflecting subjects' individual weighting of these
events. Life events scores of importance included a 6-month recall of
major events occurring just prior to enrollment, and monthly reports
of life events which were summed to produce a 6-month concurrent total
of events experienced during the investigation. The HS was also
completed monthly and summed to provide a 6-month total hassles
index. (See Appendix C for subjects' scores on all variables.)
Consistent with previous investigations (Zimmerman, 1983),
Pearson product-moment correlations of frequency and impact scores
resulted in extremely high correlations. The correlation between the
6-month retrospective LES frequency and impact scores was .95 (£=120;
£=.0001), between LES concurrent frequency and impact .95 (£=124,
£=.0001), and between concurrent HS frequency and impact .97 (£=124;
£=.0001). Although it might be argued that the weighted measures may
be more sensitive to an individual's perception of the stressful ness
of major and minor events, given the redundancy in information
provided by these indices, they were not included in further group
analyses. Simple frequency counts of negative events were used for
both the LES and HS measures as they were seen as providing the most
objective and straightforward indices of major and minor life changes,
respectively.


61
Table 1 provides means and standard deviations obtained on stress
measures for the 6 months prior to the study (past) and for the 6
months during the study (concurrent).
Table 1
Means and Standard Deviations of Past and Concurrent Stress Measures
Time Period
Stress Measure
Past 6 months
n_ = 120
Concurrent 6 months3
n_ = 117
Frequency of Major Negative
4.43
11.90
Life Events
(3.11)
(8.94)
Impact of Major Negative
9.57
21.02
Life Events
(7.16)
(17.24)
Hassles Frequency
-
121.48
(89.07)
Note. Means appear above standard deviations which are in
parentheses.
a Concurrent measures were computed by summing monthly scores over the
6-month period of the investigation.
Despite the fact that subjects dated major events for monthly
administrations of the LES to avoid overlap from month to month, the
average frequency of major negative life events is higher for the
concurrent 6-month period than for the 6-month period prior to entry
into the study, as recalled at the time of enrollment. There may be
several possible reasons for this discrepancy. First, it is possible
that the actual number of events experienced over the two 6-month time


62
periods differed. Secondly, the discrepancy may be attributed to
memory loss for events that occurred over the more remote 6-month
period prior to entry into the study (Monroe, 1982a; Paykel, 1983).
Memory loss for the concurrent period of investigation would be less
severe as recall was required over a shorter time period of 1 month.
Repeated reporting of major negative events that tend to recur
could also have contributed to the higher concurrent scores. For
example, the eight most frequently reported negative life events in
the concurrent information include events that could occur more than
once in a 6-month period. Arranged in order from the most frequent
they include sexual difficulties, major change in sleeping habits,
changed work situation, major change in financial status, major change
in eating habits, more or less arguments with spouse, breaking up with
boyfriend or girlfriend, and a major change in social activities.
These items were among those most frequently endorsed on the 6-month
recall as well. Such events could occur more often than once every 6
months and be counted a maximum of six times per subject in the
concurrent information (which is based on a total of six monthly LES
scores) but would only have been counted once on the 6-month recall of
major events experienced prior to the study.
It is of interest to compare LES scores of the present sample
with those of other groups. Normative data are provided on the LES
impact ratings of 345 male and female students enrolled in
introductory psychology classes (Sarason etal., 1978). Although
impact scores were not used in analyses in the present study, they
were available and are used here for comparison purposes since


63
normative data have been provided for LES impact but not LES frequency
scores. The mean 6-month recall negative impact score of herpes
subjects was 8.57 (^D_ = 7.16) as contrasted to a mean score of 4.66
(SD = 4.36) for males or 5.64 (SD_ = 6.43) for females for a 12-month
recall administration of the LES. One would have expected the figures
to be reversed since the data on the normative sample are based on a
time period that is twice as long (12 as opposed to 6 months). The
difference is even more striking for concurrent information with an
average frequency of major negative life events of 21.02 (SD =
17.24). The latter, as already discussed, includes information on
frequently occurring and potentially repeated major events. Authors
of the original measure suggest that the LES score of the student
(normative) population may be low relative to subjects from the
general population (Sarason et al., 1978). It appears that the
present, more heterogeneous sample has perceived relatively more
stress than the normative group.
Scores on the HS are more directly comparable to normative data
as both are based on monthly administrations (Kanner et al., 1980).
The 100 normative subjects were older, ranging in age from 45 to 64 as
opposed to the age range in the present sample of 18 to 55. The
average monthly hassles frequency score of the normative group was
20.50 which is very similar to the average for the herpes sample of
20.25.


64
Presumed Moderator Variables
Information regarding the herpes sample as a whole on arousal
seeking tendency, locus of control, and social support is provided
here for comparison purposes.
Subjects' scores on arousal seeking tendency were approximately
normally distributed with a mean of 35.02 (n_ = 120). This is slightly
higher than the average of 32.0 reported for the normative sample of
536 subjects (Mehrabian & Russell, 1974). The standard deviations of
the two samples are similar--28.58 for the herpes subjects and 29.0
for the normative group.
Locus of Control (LC) scores were also approximately normally
distributed with a mean score of 10.07 (^l = 3.87, n_ = 122). Average
scores in the literature range from 5.94 (SD_ = 3.36, n_ = 155) for
Peace Corps volunteers to 9.22 (SD_ = 3.88, n^ = 303) for college
students (Rotter, 1966). Although the standard deviations are
similar, the average score of the herpes sample is slightly higher,
indicating that, as a group, the present subjects tend to be more
externally oriented than others.
When the average Social Support Satisfaction (SSS) score is
compared to normative data, it initially appears that the herpes
subjects' SSS scores may be slightly lower. The average SSS score for
herpes subject is 141.87 (SD = 13.92, n_ = 119); for the normative
group of 602 undergraduates, the mean was 145.26 (SD_not provided;
Sarason et al., 1981). However, the distribution of scores from the
herpes subjects ranged from 2 to 213 and was highly skewed with most
of the scores grouped together at the very satisfied end of the scale


65
(median = 156.5). Subjects who were dissatisfied with their perceived
social support were not well represented in the present sample which
limits the generalizability of findings with regard to social support.
As a group, the herpes subjects appear to be slightly higher on
arousal seeking tendency and social support satisfaction and more
externally oriented than the respective normative samples described in
the literature.
As expected, major and minor events were significantly related.
The Pearson product-moment correlation of concurrent LES and HS totals
was .68 (jt_= 122; £=.0001). The HS scores were retained for analyses of
hypotheses specifically concerning minor events.
Definition of Recurrence
True herpes recurrent lesions do not always produce positive
culture results in the laboratory due to a relatively short duration
of viral shedding. Corey and Holmes (1983) report obtaining a 42%
culture positive rate from recurrent lesions. If the definition of
the dependent measure was restricted to culture-positive lesions, as
many as 58% of true episodes of herpes would be missed. If the
definition included all reported lesions, the value of laboratory
documentation in this investigation would be lost.
The decision was made to define a recurrence for the purposes of
the present study as any reported recurrence for an individual subject
for whom at least one culture-positive result was obtained during the
course of the study. This compromise was adopted to maximize clinical
sensitivity and scientific credibility. The different recurrence


66
definitions demonstrated a high degree of association with each
other. Spearman correlation coefficients of the number of recorded
recurrences and the number of positive cultures with the number of
recurrences given one positive were .66 and .71, both statistically
significant at the .0001 level. Furthermore, correlations of each of
the recurrence measures with the stress measures were very similar to
each other. (Refer to Appendix D for correlation coefficients of the
stress and recurrence measures.) Therefore, subsequent analyses
include all reported recurrences for subjects for whom at least one
positive was documented. Hereafter, unless otherwise stated, the term
"recurrences" will refer to those meeting the requirements of this
definition. During the study, 95 of the 122 total subjects reported
258 herpes lesions. Of these, 195 were cultured (76%), and 104 of
those cultured were positive (53%). This positive rate compares
favorably to that reported earlier (Corey & Holmes, 1983). There were
31 subjects whose lesions remained unconfirmed by laboratory
diagnosis.
Using recurrences as defined above, the hypotheses of the present
investigation were tested. For ease of presentation, results
concerning the existence and extent of a direct relationship between
stress and recurrences will be introduced first. Linear correlation
coefficients were used to examine this association. Following that,
analyses pertaining to the conditions under which such a relationship
exists are presented. Included here are the results of multivariate
analyses designed to investigate the combined influence of the stress
measures and presumed moderator variables. Finally, findings of


67
within-subject analyses accomplished for the purpose of exploring the
nature of the obtained relationship will be offered.
Stress/Recurrence Relationship
The first three hypotheses were tested using Spearman
correlations of stress scores and the number of herpes recurrences
experienced during the period of investigation. As defined above, all
recurrences were counted for each subject who had at least one
documented by viral culture during the 6-month study. This
information was correlated with the frequency of undesirable events
reported in the 6 months prior to the time of investigation to test
the first hypothesis. This is the purest prospective investigation of
stress and herpes recurrences since all stress information was
collected prior to the monitoring of recurrences. The result, based
on 120 individuals, was nonsignificant (£=.040; £=.656).
Because this result was surprising in light of two previous
retrospective investigations (Taylor, 1978; Watson, 1983), the
analysis was repeated for the subset of subjects who experienced
recurrences during the study, eliminating all subjects with zero
recurrences. This produced a subsample of recurring subjects, more
similar to the previous samples studied. The result was also
nonsignificant (£=-.040; £=53). These results indicate a total lack
of association between recent undesirable life events and subsequent
herpes recurrences, providing no support for Hypothesis 1.
In contrast, a significant positive relationship was obtained
between monthly totals of the frequency of reported undesirable life


63
events, summed over the 6-month period of the investigation, and
recurrences experienced over the same period of time. The correlation
obtained was .193 (£=.037; £=117). This significant positive
correlation indicates that increasing levels of reported negative life
events were associated with increasing recurrence rates when
concurrently monitored. Thus, some support was obtained for the
second hypothesis that stress and recurrences are concurrently
related.
With regard to Hypothesis 3, concerning minor life events
experienced over the same period of time, no significant relationship
was demonstrated. The correlation of the total HS scores reported
monthly and recurrences was only .019 (£=117). It appears that minor
life events are not significantly associated with herpes recurrences,
at least when monthly hassles scores are accumulated over a 6-month
period and related to recurrences over the same time period.
Influence of Moderator Variables
The fourth hypothesis concerns finding the best possible
combination of variables for predicting recurrences and evaluating the
influence of presumed moderator variables. Because it was the
particular pattern or combination of these variables that was of
interest, multivariate methods of analysis were chosen. It would have
been possible to use an analysis of variance procedure with the data
collected in this investigation in order to examine the differences in
recurrence rates associated with high or low scores on the moderator
variables and high or low frequencies of stressful events. However,


69
multivariate regression analyses make better use of the continuous
nature of the data and can provide a more precise description of the
relationships among the variables (Linn, 1982). In linear models,
multivariate techniques allow for the assessment of the effect of each
variable while statistically holding other variables of interest
constant. They also allow for nonlinear models which can include
interaction terms and therefore make it possible to assess the
moderating effects of one variable on another (SAS User's Guide,
1985).
Because the findings of the concurrent data provided the only
evidence for a stress/recurrence relationship, analyses employing
these concurrent data are the focus of the remainder of this report.
One analysis included all subjects, the second included only subjects
who experienced at least one recurrence during the study. Details of
the methods and results of these analyses and the follow-up procedures
used to verify results are described below.
When all individuals were considered in the analysis, a highly
skewed distribution of recurrences resulted due to the large number of
individuals with zero recurrences. Therefore, the recurrence variable
was categorized into three groups according to the number of
recurrences: (a) nonrecurrers who experienced no documented
recurrences during the study, (b) recurrers who had one or two
recurrences, and (c) recurrers who had three or more recurrences.
Logistic regression analysis was chosen as the most appropriate
analysis for use with this ordered categorical data (Harrell, 1983).
A logistic regression can be used to estimate nonlinear relationships


70
and takes into account the ordered nature of the recurrence categories
(Afifi & Clark, 1984; Harrell, 1983).
A forward stepwise logistic regression procedure was performed to
identify which, if any, variables and/or interactions were related to
recurrence groups. Single independent variables available to enter
the model consisted of the total number of negative major life events
reported monthly and summed over the 6 months of the study (LESNT),
the total number of hassles reported monthly and summed over the 6
months of the study (HST), and the following three variables which
were measured at the beginning of the study: locus of control (LC),
satisfaction with social support (SSS), and arousal seeking tendency
(AS). Two-factor interactions, which included at least one stress
measure (HST x LC, HST x SSS, HST x AS, LESNT x LC, LESNT x SSS,
LESNT x AS, LESNT x HST), and three-factor interactions, which
included one stress measure (HST x LC x SSS, HST x LC x AS,
HST x AS x SSS, LESNT x LC x SSS, LESNT x LC x AS, LESNT x AS x SSS),
were all available to enter the model if important in predicting
recurrences. The level of significance required for a variable to
enter and remain in the model was .05. The only term selected as
significant in this manner was the LESNT x SSS interaction (jK.OS).1
1 The logistic regression procedure was repeated with two
variations: (a) Subjects were eliminated whose cultures were never
documented by positive culture (n=31). In most cases, these cultures
were "documented" in the sense that a CRC nurse had judged the lesion
worthy of culture. Therefore, some were true recurrences for which an
attempt to isolate the virus failed, (b) Sex was included as a
variable in order to statistically control for a significant sex
difference on responses to the SSQ (_t( 117) =3.73; £=.0007). Removing
the "probable recurrers" from the group of nonrecurrers and
controlling for sex differences did not alter the logistic regression
solution.


71
A follow-up logistic regression analysis was performed with
LESNT, SSS, and their interaction as the independent variables. This
resulted in a prediction equation relating stress to recurrences. The
nature of the LESNT x SSS interaction, however, suggests that the
slope of recurrences on LESNT increases with increasing social support
satisfaction, meaning that at higher levels of SSS, there is a
positive relationship between stress and recurrences. Only those
extremely low on SSS demonstrate a negative relationship between
stress and recurrences. A change in slope occurred at approximately
100. For SSS scores greater than 100, the relationship between LESNT
and recurrences was positive, but negative for values less than 100.
In the present sample, SSS scores range from 69 to 158 with a mean of
141.2 and standard deviation of 19.8. The distribution is highly
skewed with more scores at the high end of the scale. Since less than
5% of the subjects' SSS scores fell below the critical score of 100
virtually no support was obtained for a beneficial moderating effect
of social support satisfaction on life stress. For the vast majority
of subjects, high life stress was associated with more frequent
recurrences.
Since previous investigations have focused on recurring subjects,
a regression analysis was performed using the subsample of the 52
recurrers for whom information on all independent variables was also
complete. The number of recurrences for an individual served as the
dependent variable. A square root transformation was applied to
stabilize the variance. A stepwise regression technique was utilized
in which variables which contribute significantly to the model enter,


72
remaining only if they contribute importantly in relation to all other
variables in the model (SAS User's Guide, 1985). The same independent
variables were available to enter the model as in the initial logistic
regression. The forward selection technique with a p_ level of .10
required for variables to enter and remain in the equation kept only
one term: the LESNT x LC interaction. The amount of variance
accounted for by this interaction was only 2%.
A follow-up regression analysis which included only LC, LESNT,
and their interaction was performed. This resulted in a prediction
equation which indicated that for individuals with a more external
locus of control (high values of LC), greater life stress resulted in
more recurrences. The opposite relationship held true for more
internally oriented individuals. The point at which the slope on
LESNT changed was approximately 3.43. So, for values of LC greater
than 3.43, LESNT was positively related to recurrences; for values of
LC less than 3.43 LESNT was negatively related to recurrences. In
this particular sample, LC scores ranged from 2 to 19 with a mean
score of 10.07 (SD_ = 3.87) and the distribution of LC scores was near
normal. As previously discussed, the herpes subjects tend to be more
externally oriented than normative groups (Rotter, 1966).
Follow-up regression analyses were performed separately for three
groups of individuals based upon their LC scores: (a) individuals
with LC scores between 2 and 8 (LC1), (b) those with LC scores between
9 and 13 (LC2), and (c) those with LC scores between 14 and 19
(LC3). Mean LESNT scores for these groups were 12.18, 13.48, and
13.75, respectively. The relationship between major life stress and


73
recurrences was different for these groups, as is illustrated in
Figure 2.
As can be seen by examination of this figure, the regression
lines for predicting the number of recurrences from an individual's
LESNT score are quite different in slope for individuals with the
lowest LC scores. Low LC scores are associated with belief in
personal (internal) control of events while high scores are associated
with belief in external control. Individuals with the lowest scores
(LC1) demonstrate a slightly negative nonsignificant slope, while
higher LC scores (LC2 and LC3) demonstrate a positive slope. In the
LC2 and LC3 groups, the estimated relationship was approximately the
same for both groups but only the slope for the LC2 group was
significant (£=.02). This illustrates the finding that life stress
has an important association with the probability of recurrence.
Since both regression analyses confirmed a stress/recurrence
relationship, the question of causality was raised. Does life stress
result in an increase in recurrences or vice versa? While it is not
possible to actually address the issue of causality as variables were
not experimentally manipulated in the study, a closer look at the
temporal relationship and direction of the relationship between stress
and recurrences appeared warranted. The following section is devoted
to follow-up analyses designed to suggest the direction of the
relationship between LES and recurrences.


Predicted Number of Recurrences
74
Number of Major Negative Life Events
Monthly Totals Summed Over Six Month Study
Figure 2
Predicted recurrences as a function of major negative life
events for three ranges of Locus of Control (LC) scores.


75
The Nature of the Life Stress/Recurrence Relationship
The method of data collection in this investigation allowed for a
more refined, within-subject analysis of the stress/recurrence
relationship. Information regarding life events was obtained at
regular monthly intervals, with recurrences being recorded by the date
of onset. This information was utilized in an attempt to clarify the
directionality of the stress/recurrence relationship.
Specifically, the hypothesis was that stressful events would
precede recurrences and this would be reflected by an increased number
of stressful events in the month prior to that of a recurrence. If on
the other hand, recurrences resulted in increased stress, this would
be reflected in an increased number of stressful events in the month
following the recurrence. To investigate this hypothesis, stress
difference scores were computed and summed for subjects who reported
recurrences. The difference scores were derived by subtracting the
stress score for the month following the month of a recurrence from
the stress score reported the month prior to a recurrence. If the
mean of these individual difference scores was significantly different
from zero in a positive direction, it would indicate that stress is
more likely to precede; than follow a recurrence and strengthen the
nature of the causal inference to be made. A nonsignificant or
negative result would weaken the case that stress causes recurrences,
perhaps suggesting that recurrent episodes of herpes result in
increased stress or that some third factor is related to stress and
recurrences causing similar changes to both.


76
Since it is possible that stress scores during months of a
recurrence could be inflated due to factors related to the recurrence,
these months were avoided in the analysis. That is, scores used were
from the first clear month (month without a recurrence) prior to a
recurrence and the first clear month following. Thus, if a person had
continuous recurrences in months 2, 3, and 4, the difference score was
computed between months 1 and 5. Difference scores on the LES could
not be obtained for recurrences falling in the first month since a
prior monthly stress score was not available. Similarly, recurrences
falling in the last month lacked stress information from a clear month
following the recurrence. In the case of more than one difference
score per subject, the scores were averaged for that subject. In the
event of recurrences separated by only 1 month, the month between
could be considered both a month prior and a month following.
Therefore, the month prior to the first recurrence and the month
following the last recurrence were used to compute difference
scores. Multiple recurrences within 1 month were treated as one
recurrence for the month.
The application of this procedure to the frequency of major life
events reported on the, LES on a monthly basis resulted in an average
difference score across subjects of 0.073. A one-sample _t-test for
this subsample of 55 subjects resulted in a jt-value of 0.28 which is
not significant (=.78). Because this was a within-subjects analysis
and weighted scores are more sensitive than simple frequency scores to
an individuals' perception of the stressfulness of events, this
analysis was repeated using impact scores. A similar nonsignificant


77
difference score was obtained. The average impact difference between
the month preceding and the month following a recurrence was 0.255
(_t_=0.48; £=.63). These results together indicate that the frequency
and impact rating of stressful events preceding a recurrence are not
significantly different from the frequency and impact ratings of
stressful events following a recurrence. This does not resolve the
issue of directionality.
Because some of the negative difference scores were large and
could have masked the fact that stress more often precedes
recurrences, another test was performed. A simple count of negative
and positive difference scores was obtained and tested for
significance with the McNemar test for changes (Siegel, 1956).
Twenty-one subjects reported more stress preceding a recurrence, 17
reported more stress following. The rest of the subjects had equal
prerecurrence and postrecurrence scores. McNemar's test with a
correction for continuity resulted in a Chi-square of 0.237 which is
not significant at the .05 level. Thus, neither of the within-subject
analyses provided support for the notion that negative major life
stress is a significant factor in precipitating herpes recurrences.


CHAPTER FOUR
DISCUSSION
The major finding of this investigation was a significant
positive correlation between concurrent major life events and
documented recurrences of genital herpes. Subjects reporting more
stressful events over the 6-month period of study experienced more
recurrences. The predicted prospective relationship between stress
and recurrences was not confirmed by the data. Neither information
from the total sample nor from the subset of recurring subjects showed
any relationship between stressful events prior to the study and the
subsequent experience of herpes recurrences. Minor events, or
hassles, were not related to recurrences in any of the analyses.
The influence of several presumed moderators of life stress was
also investigated. Results differed depending on whether the total
sample or a subset of recurring subjects was considered. In the
former case, when information from all subjects was considered, an
interaction between social support and major negative life events was
found to be most predictive of recurrences. This interaction
indicated that high life stress was associated with more frequent
recurrences and that the effect of life stress became greater as
satisfaction with social support increased. This result was in direct
contradiction to prediction. In the latter case, when information
from only recurrers was analyzed, an interaction between major stress
78


79
and locus of control emerged as an important predictor of
recurrences. This finding was consistent with expectation. It
indicated that major stresses were associated with concurrent
recurrences for most subjects, excluding only those at the extreme
internal end of the locus of control distribution. Thus, an external
locus of control orientation and the experience of major stressors was
associated with higher recurrence rates. Arousal seeking tendency
failed to demonstrate any significant relationship to recurrence rate
in either analysis. In summary, with respect to those moderator
variables under study, only locus of control orientation interacted
with stress as predicted. Social support satisfaction had a
moderating influence directly opposite to that expected, and arousal
seeking tendency emerged as unimportant when other variables were
consi dered.
In the context of the present stress/illness literature, the
finding of a significant association between undesirable major life
events and an illness outcome measure is not new. It is important,
however, to have replicated this finding in an investigation designed
to avoid methodological flaws of earlier studies. In the present
study, stress elevations cannot be attributed to recall bias or a need
to justify illness (Rabkin & Streuning, 1976). Monthly reporting of
stressors was designed to avoid distortion due to forgetting (Funch &
Marshall, 1984; Monroe, 1982a; Paykel, 1983). Episodes of herpes,
objectively documented by viral culture, cannot be attributed to a
need to seek attention in the face of stress (Mechanic, 1975; Minter &
Kimball, 1978) or a lowered tolerance for discomfort due to stress


80
(Harney & Brigham, 1984). Therefore, the finding of a significant
association, albeit weak, is nonetheless important. The present study
employed stringent control measures and confirmed the findings of
previous studies. This enhances the credibility of earlier findings
and strengthens confidence in the existence of a stress/illness
association.
With regard to genital herpes recurrences in particular, the
finding is even more important. As reviewed earlier, the belief in a
stress/recurrence link is based entirely on anecdotal reports (Adams
et al., 1980; Blank & Brody, 1950; Janicki, 1971; tlllman, 1947),
survey research (Bierman, 1982; "Help membership HSV survey," 1981),
and retrospective investigations (Taylor, 1973; Watson, 1983).
Prospective investigations have either not found a significant
association, possibly due to inadequate methodology (Katcher et al.,
1973) or failed to systematically monitor stress (Freidman et al.,
1977; Luborsky et al., 1976). The best empirical evidence for a
stress/recurrence association has been provided by Taylor (1978) and
Watson (1983) in retrospective studies based on self-reported,
undocumented episodes of recurrent herpes. Taylor (1978) found that
women who reported a h,igh rate of recurrence also reported a higher
frequency of stressful events over the previous year. Watson's (1983)
replication of the finding in a larger sample of males and females
strengthens confidence in it. The correlation between major negative
life events and recurrences found in the present study (r_=.19) is in
agreement with (significant and positive), but weaker than, those


31
found by Watson (1983; r=.32, r=.41). However, it is based on
evidence less subject to confounds than prior research.
The degree of confidence one may have in the findings of a
particular investigation can be related to the type of evidence it has
provided. One can conceptualize at least three levels of evidence
related to methodology in stress/illness research. Prospective
research in which stress is evaluated prior to the occurrence of
illness provides the purest and therefore the strongest evidence.
Concurrent research in which stress and illness are evaluated over the
same time period is somewhat weaker. The significant findings of the
present investigation are based on concurrent stress and recurrence
measures. Retrospective findings which form the bulk of evidence are
the weakest because the possibilities for distortion are highest.
Retrospective investigations are valued more for their heuristic
quality than for definitive findings.
The ability to demonstrate a prospective relationship to a large
extent depends upon the timing of event recording and occurrence of
illness. The impact that life events have may vary over different
periods of time (Sarason et al., 1982). Whether remote or recent
events affect a particular illness is also dependent upon the
characteristies of the illness. The timing of event-reporting in
life/stress research has been generally a matter of convenience or
convention. In an early investigation by Rahe and Holmes (cited in
Holmes & Masuda, 1974), physicians retrospectively reported stressful
events and health changes over a 10-year period. More recently, in
light of information regarding forgetting and distortion (Funch &


82
Marshall, 1984; Monroe, 1982a), shorter intervals have been used. A
decision regarding the optimal reporting interval to use in order to
observe a relationship could also be made considering physiological
information of the particular disease in question and related
stress/illness research findings.
With respect to genital herpes, there are reasons to look both at
long and very short latency periods. A long latency period is
suggested by Watson's (1983) research. He found a strong relationship
between stress during one 6-month period and recurrences in the
following 6-month period. The present study failed to replicate those
findings prospectively. Stresses in the 6-month period prior to the
present study were not related to documented recurrences during the 6-
month study. To investigate the possibility that stressful events
which preceded the study were related to recurrences in the first few
months as opposed to the entire 6-month period of the study, Spearman
correlations were done between LES scores for the 6-month period prior
to the study and recurrences in the first 2 months and the first 3
months of the study. These were -.13 (£=.23, £=85) and -.07 (£=.48,
£=85), respectively, indicating no significant relationship between
stress recalled for the 6 months prior to the study and recurrences in
the first 2 or 3 months of investigation. It is possible that the
earlier finding may have been influenced by retrospective bias or the
homogeneity of the sample (Watson, 1983) and resulted in a spurious
finding. More research is needed to resolve this issue.
An argument can be made on the basis of the present investigation
for a shorter latency. The significant findings were in the


83
concurrent information as opposed to the prospective information. The
attempt to use the monthly nature of data collection to examine the
essence of this stress/recurrence relationship more closely did not,
however, suggest elevated stress in the months prior to recurrences.
Although there could be many explanations for this observation, one
strong possibility is that the temporal relationship between stress
and herpes is of such brief duration that monthly measures are too
infrequent to capture the variability in stress that relates to
recurrences. If stress is causally related to recurrences and the
latency between the occurrence of a stressful event and a herpes
recurrence is a matter of 5 days or less, an event that occurs in the
beginning of the month could be related to a recurrence in the same
month. The elevation of stress that occurred prior to the recurrence
would not have been reflected in the score of the preceding month.
Similarly, if recurrences are causally related to stress and a
recurrence occurs in the beginning of the month, the resultant stress
could be reflected in LES scores in the same month and may not be
reflected in the scores of the following month. Thus, the analysis
based on monthly measures may not have been sensitive enough to detect
stress elevations that may be relevant regarding the directionality of
influence between stress and herpes.
The latency between the occurrence of stress and reactivation of
HSY may, in fact, be only a matter of days or hours. In support of
this, a latency of 1 to 2 days was reported by Herlig and Hoff (cited
in Janicki, 1971) and Ullman (1947) between emotional upset and
appearance of blisters. Direct neural stimulation such as occurs


84
during trigeminal surgery reliably results in recurrent herpes
episodes in 3-5 days in humans (Carton & Kilbourne, 1952; Pazin et
al., 1979). In an animal model, electrical stimulation of the nerve
produces herpes blisters within 3 days (Green et al., 1981).
The temporal relationship is an important consideration in the
design of life/stress studies. The present findings suggest the
possibility of a brief latency between stress and herpes. Therefore,
a recommendation for future research is to evaluate stress on a weekly
or even daily basis, while continuing to evaluate the possibility of
longer latencies. The data from more frequent monitoring can be
collapsed as desired to evaluate the impact of stress over a variety
of time intervals.
One of the aims of the current research was to enhance the
ability to predict illness outcome by evaluating minor stressors
(hassles) and presumed stress moderators as well as major life
events. There is a notion that the ill effects of major life stress
may be mediated by an increase in the number of hassles experienced
(Kanner et al., 1981). That is, the ill effects of major events
(e.g., moving) may be due to increased hassles (e.g., trouble getting
meals, being lonely, too many things to do). In the present research,
hassles did not demonstrate a direct relationship to recurrences and
the inclusion of hassles did not enhance the predictability of
recurrences in either regression analysis. In this sample at least,
the association of major negative life events and recurrences was
independent of an association with hassles.


85
With regard to the moderator variables, it was predicted that
they would interact with each other in such a way that a person's
scores on each of the measures would help to determine susceptibility
to recurrences under varying conditions of stress. This did not occur
in accordance with expectations for each measure. Of the three
presumed moderators chosen for study, only locus of control
orientation interacted with major stress as predicted. It was
hypothesized that individuals who do not feel in control of the events
in their lives would react more to stress and therefore be more likely
to have recurrences if stressed. There is support for this with
regard to adjustment outcomes (Johnson & Sarason, 1978), treatment
seeking (Manuck et al., 1975), and retrospectively reported herpes
recurrences (Watson, 1983). In general, a relationship is seen
between stress and illness only for externally oriented individuals.
The present study found support for this in the subsample of subjects
who had recurrences. No relationship was found between stress and
recurrences for individuals on the extreme low end of the scale
(internals). A significant positive relationship was found between
stress and recurrences for externals providing further support for
locus of control as a Stress moderator.
Social support has been widely researched by investigators using
a wide variety of measures. In general, either an overall beneficial
effect or a buffering role in relation to life stress has been
attributed to social support (Cohen & Wills, 1985). These beneficial
effects have been more consistently demonstrated with psychological
outcomes than with illness measures however. Watson (1983) found a


86
direct effect on genital herpes recurrences such that those high on
social support (perceived helpfulness) had significantly fewer
recurrences. It was important to replicate these findings in a
prospective investigation because retrospective bias and recruitment
procedures in the earlier study limit the generalizability of the
findings. Forty of the 46 subjects were recruited through a herpes
self-help group which may have biased the findings regarding perceived
helpfulness (Watson, 1983).
The findings of the present study with regard to social support
are not in accord with those suggesting either a beneficial main
effect or those demonstrating a mitigating effect on stress. The
nature of the moderating role uncovered for social support revealed
instead a significant positive relationship between stress and herpes
for individuals high on social support satisfaction. That means
people who reported more satisfaction with social support also had
more recurrences when under stress. This unexpected finding was
closely examined. In the present investigation, social support scores
were highly skewed. Most subjects' scores suggested that subjects
were either very satisfied or very very satisfied with support
received. It is possible that individuals in this particular sample
were predominantly very active socially and happy with that. However,
subjects may have overextended themselves socially, thus negatively
affecting their resistance to recurrences. Unfortunately, the
restricted nature of the sample with regard to social support limits
the generalizability of these findings.


87
It was considered that subjects reporting high satisfaction with
social support ¡nay have been more sexually active which, in turn, led
to more frequent recurrences of genital herpes. Because information
was available on the frequency of sexual activity on a monthly basis,
this possibility was empirically examined. Frequency of sexual
activity was found to be related to social support satisfaction
(£=.21; £=.024) but the number of recurrences was negatively related
to frequency of sexual activity (£=-.25; £=.007). This suggests that
the number of recurrences may influence the frequency of sexual
activity rather than vice versa.
It is unfortunate that the sources of support are not
identifiable on the SSQ used in this investigation. It was possible
to discern network size and perceived quality of support, but not
possible to know if the perceived sources of support were family,
friends, professionals, peers, confidante, or a significant other.
With regard to genital herpes in particular, the support of a
significant other could be more important than others.
It was also not possible to differentiate between types of
support on the social support measure. Cohen and Wills (1985) provide
a useful typology of four support resources: esteem support,
informational support, social companionship, and instrumental
support. They suggest that there must be a reasonable match between
coping requirements and available support in order for stress
buffering to occur. Relating this to the present findings, it is
conceivable that respondents' reported perceived satisfaction on the
SSQ is more reflective of the type of support labeled social


88
companionship by Cohen and Wills (1985). Thus, despite the tendency
to be quite pleased with available support, it may not be the type of
support that assists one in dealing with increased stress. As
suggested by Graves and Graves (1985) whose measures of illness
symptoms also provided evidence contrary to a stress-buffering role
for social support, perhaps the obligatory nature of dense social
networks is maladaptive under conditions of life change.
Together with the wide range of findings regarding social support
in the literature (see reviews by Cohen & Wills, 1985; Thoits, 1982),
the present results further emphasize the complexity of the phenomenon
of social support, and the need for more specific information
concerning source and type of support.
Arousal seeking tendency was hypothesized to exert a stress
moderating role on the basis of studies dealing with adjustment
outcomes. Subjects low on arousal seeking tendency have been shown to
be more subject to psychological discomfort under high stress
conditions than high arousal seekers (Johnson et al., 1979; Smith et
al., 1973). The present investigation unsuccessfully attempted to
replicate and extend those findings with regard to a physical outcome
(genital herpes recurrences).
Although the logistic regression and stepwise regression
procedures resulted in interesting findings, only 3% of the variance
in recurrences was accounted for. Furthermore, because a different
moderator variable was chosen in each regression analysis depending
upon whether the full sample or only recurrers were utilized, it
suggests the effects may not be robust. Social support was important


89
only when predicting who will or will not have recurrences, locus of
control was important when predicting how many recurrences the
recurrers would have.
Thus, the present investigation has provided support for a
significant but weak association between stress and illness. This
raises questions of how to proceed in future investigations of the
nature of recurrences and their relationship to stress. In order to
enhance predictability as originally planned, further methodological
refinements will have to be accomplished. The use of more frequent
stress monitoring to evaluate a variety of impact intervals and a more
specific measure of social support have already been recommended.
On the basis of the suggestive results of the present study
concerning the effects of social support and locus of control,
continued exploration of individual differences appears warranted.
Other psychological factors might also be considered. While it was
inferred that the three characteristics chosen for study in the
present investigation would enhance one's ability to cope with stress
and thus mitigate the illness-producing effects of stressful life
events, coping skills such as problem-solving ability could be more
directly evaluated. ¡
In addition to psychological factors, the role of biological
factors should be examined. Results of an earlier prospective study
of oral herpes suggest that biological factors are better predictors
of oral herpes recurrence than psychological factors including mood
and social assets. Disease history and illnesses accounted for 80-902
of explained illness (Friedman et al., 1977). A measure of illness


90
history and predisposition could potentially improve predictability.
Because of the interest in the mechanism of recurrence, it might also
be useful to evaluate hormonal and immunological parameters to examine
fluctuations which relate to stress and recurrence onset. To maximize
the probability of observing recurrences, high recurrers could be
evaluated in a single-subject design with frequent stress and
physiological monitoring.
The present findings regarding major negative life events and
recurrent herpes provide limited empirical support for the long held
notion of a stress/recurrence link. Because this evidence is
consistent within the study and based on concurrent findings, it is
stronger than that of previous retrospective findings. However,
despite stringent control measures, 96-97% of the variance with regard
to genital herpes recurrences remains unexplained. It behooves
researchers and clinicians to collaborate with professionals in an
interdisciplinary manner to further untangle the complexities of
recurrences.


APPENDIX A
STRUCTURED INTERVIEW
Initial Structured Interview
Demographic Information
1. Sex:
(1) F
(2) M
2. Age:
3. Marital Status:
(1) Single
(2) Living with significant other
(3) Married
(4) Separated
(5) Divorced
(6) Widowed
4. How many children do you have?
(1) 0
(2) 1
(3) 2
(4) 3
(5) 4
(6) 5 or more
5. What level of education have you completed?
(1) Grade school
(2) High school graduate
(3) Some college
(4) College graduate
(5) Some graduate work
(6) Master's degree
(7) Ph.D., M.D., or other advanced degree
6. Are you presently a student?
(1) Yes
(2) No
91


92
7. Have you had any major illnesses over the past 6 months?
(1) Yes
(2) No
a. Do you have any major illnesses or chronic conditions?
b. Are you taking any medication?
8. What prompted you to volunteer for this study?
9. When did you first experience problems with genital herpes?
10. When was genital HSV first diagnosed?
By whom?
How?
11. When was your most recent HSV recurrence?
12. How many HSV recurrences have you had in the last year?
None 1-3 4 or more
13. What do you think causes recurrences for you?
14. Do you at present have a regular sex partner? Yes No
15. Does your partner have herpes? Yes No
With respect to the last 6 months, what has been your usual frequency
of:
16. Geni tal/geni tal contact?
17. Oral/genital contact?
18. Masturbation?
19. With approximately how many different partners have you had
sexual contact over the past 6 months?
20. Have you had sexual contact with any partners who have genital
HSV over the past 6 months?


Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID ETL5NSS3G_RNOTS4 INGEST_TIME 2011-11-02T14:54:03Z PACKAGE AA00004862_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES


UNIVERSITY OF FLORIDA
3 1262 08554 3964


LIFE STRESS: IMPACT ON
GENITAL HERPES RECURRENCES
BY
EMILY FRANCK HOON
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1986

ACKNOWLEDGMENTS
This project involved the time, talent, effort, and cooperation
of many individuals. Because I appreciate their contributions and
want to recognize them and thank them, I address the following:
Ken Rand—patient, enthusiastic, humorous, encouraging,
knowledgable, accessible—friend: contributed freely—remembered with
warmth forever!
Jim Johnson—guidance, sense of organization, perspective,
calming presence, wisdom—a role model deserving emulation—my sincere
thanks.
Randy Carter and his assistant statisticians who helped make
sense of voluminous computer printouts.
And to the rest of the best committee a Ph.D. candidate ever
had: Eileen Fennell, Rudy Vuchinich, and Nancy Norvell.
Wendy and Robin Hoon—young people grown independent, before my
eyes, out of necessity while coping with the vagaries of their Mom's
graduate school schedule—I love both of you, and I thank you for your
help and cooperation.
Ernest Franck, my dad, who has always encouraged me to follow my
curiosity and to accept challenges.
The late Emily Franck, my rnom, who helped me to appreciate the
value of social relationships.

Pete Hoon, my ex-husband, with whom I shared my early research
experiences and with whom I continue to enjoy a satisfying
collaborative relationship as a professional and coparent.
Rob Martin, my present husband, who helped me in uncountable ways
to tie up loose ends and who provides inspiration for getting the most
out of work time and play time.
Subjects—courageously sharing personal information to advance
knowledge regarding herpes.
And the following cheerful helpers: research assistants,
laboratory personnel, and Clinical Research Center staff.
i i i

TABLE OF CONTENTS
Page
ACKNOWLEDGMENTS ii
ABSTRACT vi
CHAPTERS
ONE LITERATURE REVIEW 1
Stress and Illness ...1
Moderator Variables 4
Social Support 4
Locus of Control 7
Arousal Seeking 9
Methodological Issues 12
Retrospective Design 12
Measures of Illness 14
Confounding of Events and Outcome Measures 15
Value and Impact of Events 16
Minor Events 17
Recurrent Genital Herpes as a Model 18
Clinical Nature of Herpes Simplex Virus 18
Recurrences 19
Present Lack of Cure 21
Laboratory Documentation 21
Endogenous Pathogen 22
Stress and Herpes 23
Anecdotal Reports 23
Survey Research 25
Retrospective Research 26
Prospective Research 28
Summary 31
Theoretical Framework 33
Perception of Stressful Events 33
Proposed Physiological Mechanisms 36
Present Investigation 44
Hypotheses 45
TWO METHODS 48
Research Design 48
Sample Selection 49
i v

Measures 50
Stress Measures 50
Life Experiences Survey (LES) 50
Hassles Scale (HS) 51
Moderator Variables 52
Locus of Control (LC) 52
Arousal Seeking (AS) 53
Social Support Questionnaire (SSQ) 54
Physiological Measures 55
Screening procedure: HSV antibody 55
Dependent variable: Virus isolation and
transport 55
Procedure 56
Initial Contact 56
Second Contact 57
Continuing Contact 53
Final Contact 58
THREE RESULTS 59
Sample Characteristics 59
Undesirable Life Events 60
Presumed Moderator Variables 64
Definition of Recurrence 65
Stress/Recurrence Relationship 67
Influence of Moderator Variables 68
The Nature of the Life Stress/Recurrence Relationship 75
FOUR DISCUSSION 78
APPENDICES
A STRUCTURED INTERVIEW 91
8 SUBJECTS' TOTAL SCORES ON STRESS, MODERATOR, AND
RECURRENCE VARIABLES 94
C SPEARMAN CORRELATION COEFFICIENTS AND SIGNIFICANCE
LEVELS FOR REPORTED STRESS FREQUENCY AND RECURRENCE
DEFINITIONS 98
REFERENCES 99
BIOGRAPHICAL SKETCH Ill
v

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
LIFE STRESS: IMPACT ON
GENITAL HERPES RECURRENCES
BY
EMILY FRANCK HOON
May, 1986
Chairman: James H. Johnson
Major Department: Clinical Psychology
The present study of the effects of major and minor life events
on recurrences of genital herpes addressed the need for prospective
research with objective indices of illness (i.e., viral culture) and
provided empirical information regarding a stress/recurrence link.
For 6 months, 122 HSV seropositive subjects monitored stress and
had all recurrent lesions cultured. Subjects were administered the
Life Experiences Survey (LES) upon enrollment regarding major events
experienced over the past 6 months. This information was analyzed to
determine the extent of a prospective relationship to subsequent
recurrences. Thereafter the LES was administered monthly with the
Hassles Scale (HS) which concerns minor stressors. To identify
personal characteristics which might be related to frequent
recurrences under conditions of stress, the following presumed
moderator variables were examined: social support, locus of control
orientation, and arousal seeking tendency.
vi

In accordance with the first hypothesis, correlational procedures
provided evidence for a positive association between the frequency of
major negative life events and the number of HSV recurrences during
the study. This relationship was significant for concurrent LES
scores, but not when initial LES scores were considered in a
prospective manner to subsequent recurrences.
A logistic regression procedure using all subjects provided
evidence for a moderating influence of social support satisfaction,
the nature of which was contrary to expectation. A significant
positive stress/recurrence relationship was found only at high levels
of satisfaction. As hypothesized, information from recurring subjects
(jr=49) submitted to a regression analysis revealed a significant
positive stress/recurrence relationship only for externally oriented
individuals. Neither arousal seeking nor hassles were significantly
related to recurrences in any analysis.
Within-subject analyses were performed to examine the nature of
the concurrent stress/recurrence relationship. Stress was not found
to be significantly elevated prior to recurrences. This suggests
either a shorter latency exists between stress and recurrence onset
than could be detected; using monthly evaluations or a reciprocal
relationship exists in which stress may cause or be caused by
vii
recurrences.

CHAPTER ONE
LITERATURE REVIEW
Stress and Illness
The notion of a link between psychosocial factors and illness is
not new. Physicians have prescribed environmental and behavioral
interventions for years based upon their clinical experience and
intuition. What is new is a flurry of activity designed to document
and clarify this relationship.
With regard to life stress as a psychosocial factor in the
etiology of illness, Holmes and Rahe (1967) can be credited with
initiating a new wave of research with the development of their
measure, the Schedule of Recent Experience (SRE). This was the first
attempt to quantify, using a simple, 43-item questionnaire format, the
stressful life events believed to heighten vulnerability to illness.
The SRE and various modifications of this measure have been
widely used to provide empirical evidence for the relationship between
life stress (defined ip terms of major life changes) and illness.
Consistent but modest correlations have been confirmed with a variety
of illnesses and illness-related measures in the past two decades.
Elevated life stress has been shown to be associated with higher
reported illness rates among many groups (Holmes & Masuda, 1974; Marx,
Garrity, & Bowers, 1975; Rahe, Mahan, & Arthur, 1970) the seriousness
of illness (Wyler, Masuda, & Holmes, 1971), athletic injuries
1

2
(Bramwell, Masiida, Wagner, & Holmes, 1975), work-related accidents
(Levenson, Hirschfeld, Hirschfeld, & Dzubay, 1983), and the
complications of pregnancy and childbirth (Gorsuch & Key, 1974; Newton
& Hunt, 1984; Nuckolls, Cassel, & Kaplan, 1972; Rizzardo et al.,
1982). Illness-related behaviors such as the seeking of medical
attention and taking disability days have also been associated with
higher scores on life change indices (Gortmaker, Eckenroade, & Gore,
1982; Liao, 1977; Miller, Ingham, & Davidson, 1976; Murphy & Brown,
1980; Tessler, Mechanic, & Dimond, 1976; Thurlow, 1971).
Similarly, the onset of particular diseases has been associated
with elevated life change stress scores. These include myocardial
infarction (Connolly, 1976; Theorell & Rahe, 1971), sudden cardiac
death (Rahe & Lind, 1971), childhood cancer (Jacobs & Charles, 1980),
gastrointestinal disorders (Craig & Brown, 1984), appendicitis (Creed,
1981), tuberculosis (Hawkins, Davies, & Holmes, 1957), mononucleosis
(Kasl, Evans, & Niederman, 1979), hypertension (Lai, Ahuja, &
Madhukar, 1982), menstrual abnormalities (Harris, 1984; cited in
Creed, 1985; Siegel, Johnson, & Sarason, 1979), chronic yeast
infections (Williams & Deffenbacher, 1983), streptococcal infections
(Meyer & Haggerty, 1962), and experimentally induced cold infections
(Totman, Kiff, Reed, & Craig, 1980). What is striking in these
examples is the wide range of health-related problems which bear a
significant relationship to measures of life stress. It is as though
excess stress increases one's vulnerability to illness in general, not
just a few specific disorders. Indeed, this is consistent with the
views of Holmes and Masuda (1974) that life stress results in illness

3
through increased susceptibility to a range of health-related
problems.
In addition to studying the relationship between life stress and
illness onset, other studies have focused on the relationship between
life stress and fluctuations of health status in individuals with
chronic illness. Children with a mixed group of chronic conditions
(e.g., diabetes, asthma, blindness) who attended a 3-week summer camp
were observed by Bedell, Giordani, Armour, Tavormina, and Boll
(1977). A relationship between self-reported life change and
frequency of illness-related symptomatic episodes was found.
Similarly, asthmatic individuals monitored prospectively were found to
require more medication under conditions of increased life stress
(de Araujo, Van Arsdel, Holmes, & Dudley, 1973). Diabetic control has
also been investigated. Bradley (1979) found life stress to be
related to increases in blood sugar, the need for clinic visits, and
necessary prescription changes, particularly for insulin-dependent
diabetics. Similarly, insulin-dependent diabetics were found to be
more vulnerable to life stress than age- and sex-matched noninsulin
dependent diabetics and matched nondiabetic controls (Linn, Linn,
Skyler, <$ Jensen, 1983). Insulin-dependent diabetics perceived more
stress and showed poorer metabolic control despite similar compliance
with the medical regimen. They also demonstrated decreased immune
responsivity (Linn et al., 1983). Jacobson, Rand, and Hauser (1985)
studied the impact of life events on the long term complications of a
group of insulin-dependent diabetics. A small but significant
relationship was found between negative life events and glycemic

4
control for all subjects. In the subgroup of diabetics with
proliferative retinopathy of recent onset, a stronger relationship was
found between lack of glycemic control and the occurrence of negative
life events (Jacobson, Rand, & Hauser, 1985).
Moderator Variables
Despite the consistent finding of statistically significant
correlations, the correlation coefficients are of low magnitude,
typically less than .30, which suggests that life events measures
typically can only account for 9% or so of the variance in illness
measures (Rabkin & Struening, 1976). To improve upon this situation,
the life events measures have been administered in conjunction with
other measures believed to assess moderators of life stress.
Constitutional predisposition (Kobasa, Maddi, & Courington, 1981),
social support (Nuckolls et al., 1972; Schaefer, Coyne, & Lazarus,
1981), locus of control (Toves, Schill, & Ramanaiah, 1981), hardiness
(Kobasa, 1979), physical fitness (Roth & Holmes, 1985), and a tendency
to engage in arousing situations (Johnson, Sarason, & Siegel, 1979)
have been shown to moderate the effects of life stress.
Social Support
A frequently researched moderator variable has been social
support. This refers to the degree to which individuals have access
to social resources. Resources depend upon the relationships people
have to spouse, family, friends, neighbors, community groups, and
social institutions (Johnson & Sarason, 1984). Social support has

5
been measured in a variety of ways, from simply confirming the
availability of a confidante (Brown, Bhrolchain, & Harris, 1975) to
administering a 48-item inventory with four subscales to evaluate
esteem support, instrumental support, social companionship, and
informational support (Cohen & Hoberman, 1983).
In a recent comprehensive review of over 40 research articles,
Cohen and Wills (1985) found evidence consistent with both an overall
beneficial effect of support on well-being and a buffering
(interactive) role in relation to life stress. In the former, health
benefits are perceived to derive directly from a supportive network.
In the latter, social support is perceived to play a moderating role,
to be related to well-being only for persons experiencing the adverse
effects of stressful events. That is, in the face of major life
stress, an adequate social support system may serve to protect the
stressed individual from a pathologic outcome. Authors point out that
social support may simultaneously exert both a direct and a buffering
effect (Cohen & Wills, 1985).
Social support was found to exert buffering or positive main
effects on most of the psychological distress variables reviewed, such
as depression, loneliness, and anxiety. In fact, in one investigation
(Lin, Ensel, Simeone, & Kuo, 1979), the contribution of social support
to predicting psychiatric symptoms was more important than stressful
life events. However, findings regarding physical symptomatology have
been more variable. A clear, consistent link between social support
and decreased mortality has been demonstrated (Berkman & Syme, 1979),
but studies utilizing other health outcomes are less clear. They are

mixed when the health outcome is self-reported symptomatology. When
more stringent health measures have been used such as clinical
diagnosis and alterations in physiological functioning, a beneficial
role for social support has not always been consistently demonstrated
(Cohen & Wills, 1985).
A few examples will be given to illustrate the diversity of
outcomes regarding social support. Evidence consistent with a stress¬
buffering hypothesis comes from an early study of pregnant women by
Nuckolls et al. (1972). Those experiencing high levels of stress and
low levels of social support experienced almost three times as many
pregnancy and birth complications as their similarly highly stressed
peers who had high levels of social support. Similarly, adult
asthmatics in a high life stress-low social support group required
significantly more medication to control symptoms than their peers
with either less stress or more social support (de Araujo et al.,
1973). These examples demonstrate a protective, positive role for
social support with respect to certain physical outcomes.
In spite of the intuitive appeal of a positive effect of social
support, and in contrast to the above studies, Norbeck and Tilden
(1983) found no evidence for either a main effect or buffering role
with respect to pregnancy complications. They did, however, find a
positive main effect of emotional support on psychological
symptomatology (Norbeck & Tilden, 1983). New Zealand researchers
(Graves & Graves, 1985) failed to find the hypothesized buffering
effect of social support on self-reported symptoms of illness in three
distinct ethnic groups. In fact, of the correlations designated large

7
enough to be of clinical significance, all were in direct opposition
to the buffering hypothesis. It is suggested that the obligatory
reciprocity of dense social networks may be less helpful under
conditions requiring adaptation to life change (Graves & Graves,
1985). An interesting cyclical pattern of stress and social support
was demonstrated by Canadian investigators in a longitudinal study
(McFarlane, Norman, Streiner, & Roy, 1983). They found that help from
social support networks reduced exposure to stressful events.
However, increases in stressful events led to a reduction in perceived
he1p. No association was found between social support and physical
symptom reporting or physician visits (McFarlane et al., 1933).
Obviously, the relationship between social support, stress and
health is a complex one. Critics suggest the use of longitudinal
designs (Cohen & Wills, 1985; Thoits, 1982) and the measurement of
satisfaction (perceived quality) with available support as well as the
amount (quantity) of social support available (Sarason, Levine, &
Sarason, 1982) to further unravel the role of social support with
respect to major life events and health status.
Locus of Control
í
Another measurable variable that has shown promise as a moderator
of life stress is an individual's perception of the controllability of
situations. People vary in the degree to which they attribute the
responsibility for events to themselves or to fate. It is reasonable
to assume that individuals who perceive themselves as having little or
no control would feel more threatened in the face of undesirable life

3
events than those who feel capable of influencing outcomes. A more
intense adverse reaction could contribute to a more severe
physiological response to life stress.
The Rotter (1966) Locus of Control Scale measures the degree to
which individuals perceive themselves as having control over naturally
occurring life events. It has provided a tool for investigation of
perception of control, life stress, and psychological and physical
outcomes. Individuals are classified as internals on the scale if
they believe in personal responsibility and externals if they
attribute the control of events to chance. Johnson and Sarason (1978)
administered the Locus of Control Scale along with a major life events
scale and measures of depression and anxiety to college students.
They found negative life changes to be significantly related to both
trait anxiety and depression for externals, but not for internals. A
similar relationship was found for males in a study of life stress,
locus of control, and health status as measured by the Cornell Medical
Index (Toves et al., 1981). Negative life change was related to
health status only for external males. Females demonstrated a
relationship between negative events and illness regardless of
perception of control (Toves et al., 1981). In further support of the
notion of perceived control as a moderator of life stress, Kobasa
(1979) demonstrated that individuals with an internal locus of control
may be protected from adverse physical consequences of major life
change. Compared to business executives with high stress/high illness
scores, those with high stress/low illness indicated a more internal
locus of control. In this investigation, perception of internal

9
control emerged as one component of a constellation of factors named
"hardiness" (Kobasa, 1979).
The above studies are all retrospective in nature, and it could
be argued that individuals who have experienced recent illness would
see themselves as victimized and thus tend to respond to the locus of
control scale as externals because of their poor health status. To
strengthen the causal inference that hardiness (which includes an
internal locus of control orientation), Kobasa (1981) replicated her
findings in a prospective design suggesting that an internal locus of
control is related to resistance to illness under conditions of
stress. Manuck, Hinrichsen, and Ross (1975) followed college students
for 6 months after assessing locus of control orientation and life
stress. Although they found no difference in illness-related
treatment-seeking behavior between internals and externals under high-
stress conditions, more low-stressed externals sought treatment than
low-stressed internals (Manuck et al., 1975). Passer and Seese
(1983), however, failed to prospectively demonstrate a significant
moderating effect of perception of control with athletic injury as the
dependent variable. These findings suggest that locus of control
merits further investigation as a moderator of the effects of major
life stress.
Arousal Seeking
Individuals differ in the degree to which they seek out or
attempt to avoid arousing situations. Inventories have been devised
to evaluate this tendency enabling researchers to investigate this

10
characteristic as a moderator of life stress (Mehrabian, 1978;
Mehrabian & Russell, 1974; Zuckerman, 1971). It is hypothesized that
those with a low optimal level of arousal, who tend to avoid
stimulation (low arousal seekers), would be more likely to respond
negatively to undesirable life events than those who have a high
optimal level of arousal and thus seek out stimulation (high arousal
seekers). High arousal seekers are presumed to be better able to deal
with the increased arousal brought about by the experiencing of life
changes (Johnson & Sarason, 1984).
This prediction has been addressed by a few researchers. Smith,
Johnson, and Sarason (1978) found negative life change to be
significantly related to discomfort scores on the Psychological
Screening Inventory for only the college students low on a sensation
seeking measure. Extending these findings using a different measure
of arousal seeking and measures of anxiety, depression, and hostility,
Johnson et al. (1979) found negative change to be related to anxiety
and hostility only for low arousal seeking subjects. At least one
researcher (Cohen, 1982) has failed to replicate the above results.
These preliminary findings with regard to psychological outcomes
suggest that a moderating effect of arousal seeking tendency on life
stress with regard to certain physical outcomes may also be worth
investigation. Individuals low on the arousal-seeking dimension may
be much more likely to experience physiological reactivity in the face
of life stress than others, which could adversely affect general
heal th.

11
In summary, as a means of investigating the variability in
individual susceptibility to health status changes in the face of
undesirable life stress, the investigation of the roles of moderator
variables appears to be warranted. The potential influence of social
support, locus of control, and arousal seeking tendency are of
particular interest based on findings to date.
With the exception of the Kobasa investigations (Kobasa, 1979;
Kobasa et al., 1981), moderator variables have been evaluated singly
regarding their potential to alleviate or exacerbate the negative
effects of life stress. More information is needed on how these
variables interact or combine in an additive fashion to influence
health status in the face of life stress. Multiple regression
techniques provide an analytic tool appropriate for clarifying the
relative importance of multiple potential moderator variables.
Increased documentation of significant associations between
stressful life events and the onset or course of physical illness has
changed the degree of complexity of the empirical questions. Instead
of asking vf stress is related to a particular illness, the questions
to answer now are for which individuals, under what conditions, and to
what extent are life events related to this disease? Answering more
complex questions requires more complex research designs (Johnson &
Sarason, 1984). In the next section, some of the methodological
shortcomings of present research will be discussed. Addressing these
issues will enable us to be more specific about our research questions
and to further clarify the nature of the stress/i11 ness relationship.

12
Methodological Issues
Methodological issues limit interpretations and findings in this
area of research. These include retrospective bias, inadequate
indices of illness, failure to discriminate between desirable and
undesirable events, and failure to account for minor stressful events.
Retrospective Design
The commonly utilized retrospective design relies on recall of
events and illnesses. Findings based on memory may reflect
individuals' expectations regarding stress and illness, their need to
justify illness (Rabkin & Struening, 1976), or their tendency to
report negatives due to feelings of depression (Johnson & Sarason,
1984). Retrospective reporting is also subject to forgetting as
demonstrated by Monroe (1982a). He investigated memory of major life
events utilizing prospectively collected information as a baseline to
indirectly estimate the most recent retrospectively reported four-
month period. Underreporting of events was estimated in this way to
be as high as 61$, with desirable events relatively more susceptible
to distortion with passing time than undesirable events (Monroe,
1982a). Funch and Marshall (1984) examined the rate of fall-off in
event reporting over a retrospective 30-month period. Using estimates
based on 12- and 18-month recall periods, they concluded that fall-off
is most rapid in the first 12 months, approximately 5% per month. It
then tends to level off. Fall-off was related to type of event in
their research, the more salient events showing the least fall-off.
Respondent variables such as income, education, and marriage were

13
related to the tendency to remember disruptive events. That is,
subjects with fewer resources were most likely to underreport
stressful events (Funch & Marshall, 1984). In a review of several
articles which address the issue of event underreporting, Paykel
(1983) found consistency in estimates of 4-5% loss per month in
retrospectively self-reported life events, with a lower rate (1-3%)
for event reporting by interview. To avoid retrospective bias due to
forgetting, frequent short-time intervals for reporting have been
recommended (Cleary, 1980; Monroe, 1982a), with interview techniques
used whenever possible (Paykel, 1983).
Prospective studies avoid this kind of distortion. Jenkins,
Hurst, and Rose (1979) reviewed studies on cardiac and cancer patients
finding prospective studies which contradicted or diminished the
significance of earlier findings of retrospective investigations.
Unfortunately, good prospective studies of the onset of illness are
very cumbersome and costly to implement. Many people must be
investigated who may never get the disease or diseases in question
during the course of study. For example, of 1400 cadets studied for
four years, only 194 became infected with infectious mononucleosis
during the investigatipn and, of these, only 48 developed clinical
symptomatology (Kasl, Evans, James, & Niederman, 1979). In another
example, 4,486 widowers were followed for 9 years to conclude that
there is a 40% increase in mortality within the first 6 months of
bereavement (Parkes, Benjamin, & Fitzgerald, 1969). Prospective
investigations of the effects of life stress on the clinical course
and outcome of disease are more easily and economically

14
accomplished. These involve prospectively following already
identified patients with respect to life stress along with
fluctuations in their health status regarding the illness in
question. The data from all subjects remain important for analysis,
not just those who happen to die or become infected with a particular
disease.
Measures of Illness
Another weakness of research in this area has been the failure to
employ strict measures of illness (Johnson & Sarason, 1984). Self-
reports of symptoms, the seeking of medical treatment, and medication
needs may not reflect the existence or severity of a physical
disorder. Additionally, high life stress may increase the tendency to
seek help for minor illness (Mechanic, 1975) or lower an individual's
threshold of complaint and tendency to adopt the sick role (Minter &
Kimball, 1973). Harney and Brigham (1984) experimentally demonstrated
that an individual's tendency to tolerate discomfort (cold pressor,
loud noise) is related to the amount of recent life change
experienced. This suggests that high life change individuals may be
more likely to seek aid than others for identical symptoms (Harney &
t
Brigham, 1984). One way to avoid these confounding factors is to
employ more rigid measures of illness. Rather than reliance on
illness behavior, medical records, or self-report, more objective
indices are suggested such as x-ray and other clinical and laboratory
data which provide biological evidence of the presence or severity of
disease (Minter & Kimball, 1978).

15
Confounding of Events and Outcome Measures
A concern of many in this area of research in the possible
confounding of independent and dependent variables in life stress
measures (Johnson & Sarason, 1984; Monroe, 1982b). Many items which
are considered to be stressful events may also be consequences of
illness (e.g., sexual problems, major personal injury, changes in
eating or sleeping habits). An argument can be made that such items
should be removed or analyzed separately from the domain of major life
events to avoid artificially inflating the stress scores of
individuals with illness (Dohrenwend & Dohrenwend, 1974). However,
such experiences may themselves represent significant sources of
stress. To ignore them in a consideration of the total stress of an
individual may be to neglect important information and result in a
loss of sensitivity in the life stress measure. In an investigation
of etiological factors associated with a particular disease, a
division of items into three subgroups for analysis is recommended:
(a) events that may be confounded with psychiatric condition, (b)
events consisting of physical illness and injury, and (c) events which
are independent of either an individual's health or psychiatric
i
condition (Dohrenwend & Dohrenwend, 1974; Monroe, 1932b). When the
major purpose of the investigation is to predict illness from stress
levels, it is reasonable to include all items as reflective of the
total stress experienced (Johnson & Sarason, 1984). This would be
particularly appropriate when the illness dependent variable is
specific and objective (e.g., laboratory documentation of the presence

16
of a disease organism). Confounding problems are more likely when
general, subjective measures of illness serve as the dependent
variable (e.g., reported symptoms, treatment-seeking behavior).
Value and Impact of Events
Other methodological considerations include the value (negative
or positive) and impact (weighting) of events. Until recently,
investigators have failed to discriminate between desirable and
undesirable events. In those investigations in which a distinction
has been made, negative change has proven to be more predictive of
personal maladjustment (Mueller, Edwards, & Yarvis, 1977; Vinokur &
Selzer, 1975) and physical illness (Hotaling, Atwell, & Linsky, 1978;
Johnson & Sarason, 1984; Sheehan, O'Donnell, Fitzgerald, Harvey, &
Ward, 1979).
Different methods have been explored for weighting the impact of
stressful events to obtain a more sensitive measure of individual or
consensus-derived perception of the stressfulness of particular life
events. However, raw event totals and weighted event totals are
typically highly correlated with each other. Zimmerman (1983) found
an average correlation of .94 in a review of 19 studies. Since the
I
differential weighting procedures are functionally equivalent to
unweighted procedures in group research, it is parsimonious to employ
the simplest method: a simple count of events (Johnson & Sarason,
1984; Monroe, 1982b; Skinner & Lei, 1980; Zimmerman, 1983). This does
not deny the clinical relevance of subjective weighting procedures for

17
understanding particular individuals. It simply acknowledges that
such refinements are unnecessary and redundant for group research.
Minor Events
The final methodological issue relates to the consideration of
minor as well as major life events. Because major life events per se
do not represent all stressful events, the Hassles Scale (HS) has been
developed (Kanner, Coyne, Schaefer, & Lazarus, 1981). In contrast to
major life events such as divorce, job loss, and pregnancy, the HS
deals with minor stresses, annoying practical problems such as making
mistakes, being bored, and having to wait. Already there is some
suggestion that hassle scores may be more strongly related to a
general health measure than life events scores (DeLongis, Coyne,
Dakof, Folkman, & Lazarus, 1982).
In order to account for a greater proportion of the variance in
the relationship between stress and illness, the present study was
designed to address some of the existing methodological limitations.
To avoid retrospective bias, a prospective research design with
monthly concurrent assessments of stressors was planned. The Life
Experiences Survey (LES) was chosen to measure major life stress
because it distinguishes between negative (undesirable) and positive
(desirable) events (Sarason, Johnson, & Siegel, 1978). The Hassles
Scale (HS) was utilized to evaluate minor stressful events (Kanner et
al., 1981). Additionally, three hypothesized moderator variables were
measured to evaluate their effectiveness in combination in

18
ameliorating the effects of stress. And, finally, a population was
chosen for which an objective index of illness could be obtained.
Recurrent Genital Herpes as a Model
The choice to study individuals with recurrent genital herpes
virus in this investigation was based on the nature of the illness,
the need for information concerning the precipitants of reactivations
of the virus, the ease of obtaining an objective measure of the
disease, and the availability of a motivated population of subjects.
Given these considerations, an investigation of herpes simplex virus
would seem to be a fruitful means of studying the relationship between
life stress and fluctuating health status in chronic illness.
Clinical Nature of Herpes Simplex Virus
Herpes simplex virus (HSV) types 1 and 2 are two of several
herpes viruses which affect humans and share a unique capacity to
produce lifelong latent infection despite host development of antibody
and cellular immunity. Serious complications can occur including
ocular infection, encephalitis, and neonatal infection. Fortunately,
these are rare and can usually be prevented. There also appears to be
an association between genital herpes and cervical cancer, but so far,
direct evidence of a causal link is lacking (Rand, 1982).
Neither serious complications nor significant physical
consequences are responsible for the widespread fear and attention
elicited by the disease. There is evidence that the incidence of
genital herpes is rising rapidly. Consultations to clinicians for

19
genital herpes have increased almost nine-fold in the past 20 years
(Mertz & Corey, 1984). Its incurability, transmissabi1ity, rising
incidence, and sexual nature create the concern. This can only be
understood in light of information regarding the natural course of the
disease.
Herpes is almost always transmitted by intimate contact between
infected and noninfected skin surfaces, particularly the mucous
membranes of the oral cavity and genitalia. Once inside, herpes takes
over the protein-producing apparatus of the host cell and reproduces
itself until it is brought under control by the immune system. During
an initial (primary) infection, this process produces multiple,
painful blisters or ulcerative lesions accompanied by fever, swollen
lymph nodes, and malaise. The lesions last an average of 19 days
(Mertz & Corey, 1984).
Within hours of initial exposure, the virus retreats via nerve
fibers to a safe hiding place in the trigeminal ganglia (oral herpes)
or the sacral ganglia (genital herpes) which is located near the
spinal cord. It is this capacity of the virus to reside latently
without destroying the ganglion and to periodically reactivate to
cause recurrent lesions which frustrates patients, doctors, and
researchers.
Recurrences
Almost all patients experience recurrences, from one in a
lifetime to two or three a month. The risk of recurrence when the
infectious agent is HSV-2 is higher than with HSV-1 regardless of the

20
site of the infection. Otherwise, the clinical manifestation of HSV-1
and HSV-2 is identical (Mertz & Corey, 1984). A diverse group of
factors are associated with reactivation of HSV including fever
(Keddie, Reeves, & Epstein, 1941), local irritation or tissue injury
(Blyth & Hill, 1984; Wickett, 1982), lack of sleep, overexertion, and
emotional stress (Bierman, 1982). Although there is widespread
acceptance of a stress/recurrence relationship, the evidence for it is
weak. The notion is based on anecdotal reports and a handful of
retrospective studies. The further clarification of the nature of
this relationship is important for proper disease management.
The physical consequences of recurrences are minimal. The
lesions cause discomfort but are rarely severe. They may last up to 2
weeks, but are sometimes fleeting, lasting a day or less. The patient
can usually carry out normal daily activities during recurrent
episodes. Between recurrences patients are completely well. However,
patients are highly infectious during a recurrent episode and may
transmit the disease even when asymptomatic (Mertz & Corey, 1984).
Patients find the periodic imposed abstinence from sexual activity to
be disruptive and to result in decreased sexual spontaneity. Female
patients have concerns about their reproductive capability. Single
individuals must deal with disclosure to potential partners, causing
embarrassment, shame, and fear of rejection (Luby & Klinge, 1985).
Unfortunately, because of its sexual nature, biased reporting
regarding genital herpes in the media has contributed to popular panic
about the disease, stigmatizing patients, and exacerbating the
emotional turmoil involved (Bierman, 1985).

21
Present Lack of Cure
At the present time, no cure is available for the disease. It
has recently been demonstrated that oral administration of daily doses
of acyclovir suppresses reactivation of the virus in patients with
frequently recurring genital herpes (Douglas et al., 1984; Straus et
al., 1984). However, when the prophylactic therapy ceases,
recurrences resume at the prior rate. And, unfortunately, the expense
of the drug and as yet unknown long term side effects prohibit its use
by many. Until a cure can be found, management of the disease must
consist of counseling patients to prevent transmission by practicing
sexual abstinence during recurrences and preventing recurrences.
Prevention of frequent recurrences with acyclovir is an option for
some. With better knowledge of the precipitants of frequent
recurrences of genital herpes, behavioral counseling for drug-free
prevention for all patients will improve. Thus, there is a compelling
need to learn more about the psychological and behavioral factors
associated with recurrences.
Laboratory Documentation
It is possible to obtain objective evidence of the presence of
herpes simplex virus by laboratory methods. The virus may be
cultivated in tissue culture, viral particles may be detected by
electron microscopy, or immunologic methods may be used to detect
viral antigen or cytopathologic change. The isolation of the virus in
tissue culture has been shown to be the most sensitive measure. The
virus has been successfully isolated in 82% of ulcerative lesions in

22
primary genital herpes episodes and 42» of ulcerative lesions in
recurrent episodes. The decrease in percent viral confirmation in
recurrent episodes as opposed to primary can be attributed to a
shorter duration of viral shedding in the former (Corey & Holmes,
1983). The fact that objective evidence of the presence of the
disease is possible also contributed to the choice to study this
particular group of individuals in a stress/illness paradigm.
Endogenous Pathogen
A final factor regarding recurrent genital herpes is that
extraneous exposure to an unknown dosage of an exogenous pathogen was
not an issue. Stress would be most important as a contributing factor
in the etiology of a disease when the host/microorganism relationship
is in delicate balance (Plaut & Friedman, 1981). Uncontrolled,
extraneous dosages of a pathogen may be so high that disease is
inevitable, or too low to cause disease under any conditions.
Likewise, the more virulent the microorganism, the lower the dose
necessary to introduce infection. Recurrent herpes lesions depend
upon an endogenous source of infection which remains stable within an
individual unless reinfection occurs from an outside source.
Genital herpes, therefore, provides a good model for the study of
stress/illness relationships. It is a chronic, recurring disease
which causes emotional turmoil for patients and their sexual partners
despite minimal physical consequences. Between recurrent episodes,
patients are unaffected by disease presence. Patients are thus
motivated to avoid recurrences and willing to participate in research

23
to learn more about precipitants of herpes reactivations. Viral
isolation is possible by laboratory techniques providing an objective
measure of the presence of the disease, and extraneous exposure to
pathogens of unknown dose and virulence is not a concern. Patients,
physicians, and individuals who have never had genital herpes will all
benefit from improved understanding of the disease.
Stress and Herpes
In the following section, investigations which have attempted to
clarify the relationship between stress and herpes will be
presented. Herpes simplex virus-type 2 most commonly affects the
genital area, but HSV-1 may also do so. Because they are clinically
indistinguishable (Rand, 1982), research including either HSV-1 or
HSV-2 will be reviewed.
Anecdotal Reports
Schneck (1947) reported the case of a male who could predict
HSV-1 recurrences which occurred within 24 hours of emotional
stress. These episodes were most likely to occur when appropriate
outlets for hostile feelings were unavailable. According to the
account, improved management of hostility resulted in decreased
recurrences.
Blank and Brody (1950) provided psychoanalysis for 10 patients
with recurrent oral herpes. Patients were seen weekly for 2 to 50
hourly sessions. Nine of the ten patients were described as passive,
anxious to please, dependent, immature, and hysterical. In two cases,

cancellation of an appointment by the therapist and the resultant
feeling of rejection is said to have precipitated a herpes
recurrence. From their experiences with this group of patients,
authors claim support for the use of psychotherapy in decreasing the
frequency of recurrences (Blank & Brody, 1950).
There have been reports of an association between hypnotic
suggestion and recurrent herpes episodes. Herlig and Hoff (cited in
Janicki, 1971) successfully induced oral herpes recurrences in three
female patients by reminding them of unpleasant emotional situations
and itching sensations while under hypnosis. Recurrent lesions
appeared in 1 or 2 days in all three subjects. Ullman (1947)
similarly induced an oral herpes episode in a male patient by giving
hypnotic suggestions that herpes blisters would form and that he
appeared debilitated. Within 24 hours the man had blisters on his
lower right lip which appeared to be oral herpes. Hypnosis has also
been reported to have a positive effect on the recurrence rate of
HSV-1 (McDowell, 1959).
Many supposed "cures" of herpes have been attributed to placebo
effects, or positive expectancy. One investigator reports favorable
reactions in 50% of herpes patients who received injections of steril
water (Kern, 1979). Hamilton (1980) reports even more dramatic
placebo results. In a double-blind placebo-controlled trial of ether
as a treatment for herpes, 75% of the patients on placebo experienced
improvement of herpes symptomatology (Hamilton, 1980, p. 54).
Anecdotal reports of an association between hypnosis or
expectancy and episodes of recurrent herpes are suggestive, but

25
inconclusive. Failures to induce herpes blisters through suggestion
are unreported. However, the existence of a link between emotional or
cognitive factors and recurrent herpes as can be hypothesized from
these reports is worthy of more systematic investigation.
Survey Research
Results of epidemiological survey research also suggest a link
between emotional factors and recurrences. In a national survey of
3,148 individuals with genital herpes, 83% of the respondents
acknowledged stress as a factor in recurrences ("Help membership HSV
survey," 1981). The subjects attributed the following ill effects to
having contracted herpes: periodic depressions (84%), sense of
isolation (70%), conscious avoidance of intimacy (53%), cessation of
sexual activity (10%), diminished sex drive and/or impotence (35%),
suicidal feelings (25%), dissolution of a long term marriage or
relationship (18%), rejection by a potential sex partner (21%), and
loss of self esteem (40%). Herpes is cited as both a cause and effect
of stress by these participants who were recruited through the
newsletter of a support network for individuals with genital herpes.
Two-thirds of the sample reported experiencing more than five
recurrences in 1 year. Although the sample may not be representative
of all genital herpes patients, it does reflect the perceptions of a
substantial number of patients. Another, possibly more representative
sampling of 825 patients, 53% of whom believed they were cured of
genital herpes, yielded a similar finding. In this group, emotional

26
stress emerged as the most important self-reported mechanism for
triggering recurrences, cited by 86% of the sample (Bierman, 1982).
Retrospective Research
Taylor (1978) studied life events and herpes recurrences in a
sample of 60 female students, of whom 36 had genital herpes and 24
served as controls. All were administered the Life Experiences Survey
and a Genital Herpes Questionnaire. Women with genital herpes did not
report a higher frequency of stressful events over the prior year than
women without herpes. Within the herpes groups, however, women with a
high rate of recurrence (four or more in previous year) were compared
to those with a low recurrence rate (three or less). The high
recurrers reported having significantly more negative events than the
low recurrers. Due to the retrospective nature of the study, it is
impossible to know whether high levels of negative events were a cause
or effect of herpes reactivations. In the same study, 70% of the
women indicated that having genital herpes caused emotional stress.
Additionally, 92% of the herpes patients recalled experiencing stress
the week prior to a recent occurrence, while only 42% of normal
controls recalled stress at a comparable period. The potential
retrospective bias is apparent. In addition to patients' expectations
regarding stress and illness, they had a dramatic event (herpes
episode) to tie their recollections to while control subjects may not
have had a temporally contiguous event to aid in their recall of
stresses occurring at that time. These findings are supportive of a
stress/recurrence relationship, but are limited by retrospective bias,

27
the lack of objective documentation of the illness measure, and the
restricted nature of the sample (female students).
Watson (1983) extended Taylor's (1978) findings in an
investigation of 51 male and female volunteers who had at least one
recurrence of genital herpes within the 7 months prior to
participation. A significant positive relationship was found between
undesirable life events in the previous year as reported on the Life
Experiences Survey and genital herpes recurrences in the previous 6
months. Recurrences in the prior 6 months were also significantly
related to undesirable life events reported for the 12- to 6-month
period prior to the study, suggesting a long term effect of major life
events on herpes recurrences.
Watson (1983) is the only investigator to consider potential
moderators of life stress and herpes recurrences. He included
measures of perceived locus of control and social support in his
research design. Using the Rotter Internal/External Control Scale
(Rotter, 1966), Watson (1983) demonstrated an interaction between
locus of control and stress in which subjects who perceive internal
control experienced fewer recurrences under high stress conditions
than those with an external locus of control. This interaction effect
suggests a moderating influence of locus of control on the negative
effects of stress. Social support as measured by total helpfulness on
the Social Relationship Scale did not appear to have a similar
moderating effect on stress. However, a more direct effect was
found. Subjects reporting high levels of helpfulness had
significantly fewer recurrences than those reporting low helpfulness

28
(Watson, 1983). This investigation emphasizes the importance of
investigating potential moderating factors. It is limited by reliance
on retrospective reporting and the nonrepresentative nature of the
sample, primarily high recurrers recruited from a self-help support
group.
Prospective Research
A series of prospective investigations have been undertaken to
examine the relationship of psychosocial variables and recurrences of
oral HSV (Friedman, Katcher, & Brightman, 1977; Katcher, Brightman,
Luborsky, & Ship, 1973; Luborsky, Mintz, Brightman, & Katcher,
1976). Independent measures included scores on the Life Change Index
reported retrospectively over the 2 years prior to the study, the
Cornell Medical Index, John Hopkins Symptoms Index, Clyde Mood Scale,
a Social Assets Scale, illness history, and blood analysis. Dependent
variables included illness records, infirmary visits, and, when
possible, documentation of HSV recurrences.
A year long investigation of 67 paid volunteers, 14 of whom had
significant antibodies to HSV, found previous history and blood serum
antibody presence most predictive of oral herpes recurrence, with the
unhappy factor of the mood scale also contributing significantly
(Katcher et al., 1973). When only the psychological variables were
considered, the unhappy factor accounted for 16% of the variance. The
relationship between predictors was different for herpes and systemic
illness, suggesting that recurrences do not reflect an illness
disposition. A significant negative relationship of social assets and

29
herpes episodes implies a possible moderating effect of social
assets. The 2-year retrospective life events measure was of minimal
value in all analyses and was disregarded in the subsequent
investigations (Friedman etal., 1977; Luborsky etal., 1976). While
life events were not an important variable in this study, this may be
explained on the basis of methodology. The number of subjects who
were seropositive for herpes was small. After attrition, only 33
subjects completed a full year of participation. The stepwise
multiple regression procedures were based on 19 episodes of herpes,
only 10 of which were documented by examination and viral culture.
The life events measure may have been insensitive due to memory
decline (Monroe, 1982a), the failure to consider negative events
separately (Dohrenwend & Dohrenwend, 1974; Johnson & Sarason, 1984),
or a lack of association between temporally remote events and
recurrences. However, the investigators anecdotally report an
increase of herpes recurrences during exam time for students (Katcher
et al., 1973).
A 3-month daily investigation of 20 herpes seropositive subjects
failed to demonstrate any systematic relationships between mood and
herpes reactivations (Luborsky et al., 1976). However, almost all
subjects reported stresses prior to episodes when responding
retrospectively. The authors suggest that by having moods reported in
the morning, the emotional impact of the day was missed. But it may
also be the case that moods are not necessarily equivalent to stress.
The researchers then followed a larger group of participants for
3 years (Friedman et al., 1977), during which 51 of 149 had at least

30
one HSV episode. Using herpes incidence as the dependent variable in
a multiple regression analysis, the social and psychological variables
were found to be of least predictive value (3-5% of explained
variance), biological factors (disease history and illnesses,
especially upper respiratory infection) being of greatest value
(80-90% of explained variance). Unfortunately, these investigators
did not include a stress measure. They replicated the earlier finding
that predictors were specific for herpes and unrelated to predictors
for upper respiratory infection (Friedman et al., 1977; Katcher et
al., 1973).
Daily samples of vaginal secretions from five women, three of
whom were positive for genital herpes were collected for a month by
investigators of asymptomatic viral shedding (Adam, Dressman, Kaufman,
& Melnick, 1980). Herpes simplex virus was isolated at least once
from each of the women with a history of herpes. Multiple positive
results by immunoperoxidase staining of cervical-vaginal smears were
found for all three. No statistical analyses were performed, but
graphical representation revealed temporal clusters of positive
results for each woman. These were anecdotally related to
menstruation, emotional stress, exam periods, and problems with
children. Unfortunately, the stress data were not collected in a
systematic manner making findings difficult to interpret (Adams et
al., 1980).
The data from a group of 58 psychiatric patients with primary
episodes of genital herpes who were followed prospectively for 28
weeks by Goldmeier and Johnson (1982) provide some support for a link

31
between psychosocial factors and recurrence. Subjects completed a 60-
item screening measure of psychiatric symptoms, the General Health
Questionnaire, on which they indicated which feelings and behavior
(symptoms) had been experienced over the past 4 weeks. They were then
requested to return to the clinic for viral cultures in the event of a
recurrence over the next 28 weeks. Subjects who did not return were
mailed a follow-up form requesting recurrence information. Of the 29
subjects reporting a recurrence, 7 returned for culture confirmation
of herpes. General Health Questionnaire Scores of recurrers were
compared to the nonrecurrers. The nonrecurrers had significantly
lower scores than the recurrers. Furthermore, the 29 subjects above
the cut-off score of 11 which indicates potential or overt psychiatric
problems had a significantly higher recurrence rate as determined by
actuarial recurrence-free curves. Investigators infer that anxiety or
obsessionality may increase production of adrenergic substances and
contribute to herpes reactivation. This hypothesis is consistent with
literature implicating the autonomic nervous system in decreases in
immune responsivity (Hall <& Goldstein, 1981; Rogers, Dubey, & Reich,
1979).
Summary
To summarize, despite the widespread popular notion that stress
and recurrences of herpes are linked, consistent empirical support for
such a relationship is lacking. A few anecdotal reports suggest a
connection between psychological distress and recurrences (Blank &
Brody, 1950; Schneck, 1947) and hypnotic suggestion is reported to

32
influence recurrences (Janicki, 1971; Ullman, 1947). Strong placebo
effects have been apparent in controlled investigations (Hamilton,
1980; Kern, 1979). Survey research demonstrates that patients with
recurrent herpes believe there is a connection with emotional stress
(Bierman, 1982; "Help survey," 1931). In retrospective investigations
of patients with genital herpes, high recurrence rates have been
associated with high major life stress (Taylor, 1978) and, in one
study, negative life stress appears to have exerted a long term effect
on recurrence rate (Watson, 1983). Unfortunately, anecdotal reports,
survey research, and retrospective studies are all subject to
distortion.
In one prospective investigation of oral herpes recurrences, a
negative mood factor contributed significantly to recurrence rate, but
a life events measure proved to be unimportant (Katcher et al.,
1973). Moods failed to hold up as important in subsequent prospective
studies in which life stress was unfortunately not measured (Freidman
et al., 1977; Luborsky et al., 1976). Anecdotally related stress
episodes were graphically associated with laboratory evidence of viral
shedding for three women (Adams et al., 1980). And finally, patients
demonstrating a strong potential for psychiatric problems on a
screening device subsequently reported more recurrences than those
falling below the critical cut-off score, suggesting a connection
between anxiety and recurrences (Goldmeier & Johnson, 1982). So far,
empirical support for a stress/herpes association is suggestive rather
than strong. A longitudinal prospective study with regular,

33
systematic collection of stress data and laboratory documentation of
recurrences would contribute valuable information.
In the following section, an attempt will be made to argue for
the significance of such a study in the broader scheme of
health/illness research. A reasonable proposal will be made of the
sequence of environmental, psychological, and physiological events
which could take place if stress is, indeed, a significant factor in
precipitating herpes recurrences.
Theoretical Framework
Perception of Stressful Events
Between an environmental event and a recurrence of genital herpes
many levels of complex interactions are possible. Figure 1
illustrates in simplified form the hypothesized process by which
environmental events could result in recurrences of genital herpes.
First, there must be a stimulus, the occurrence of a major or minor
life event. For it to be reported as a stressor or a hassle, it must
be perceived as such. Moderating influences may affect an
individual's perception of events in either negative or positive
ways. A person who craves excitement would probably view the event of
a flat tire enroute to work differently than a counterpart who thrives
on routines and predictability. A person who feels cared for and
supported by family and friends will probably be less threatened by
criticism on the job than a more solitary individual. The point is
that the same event could be perceived as a stressor by one
individual, but not another.

Simplified model of some of the hypothetical levels through which environmental events may
affect biological events and ultimately result in a herpes recurrence.


36
Once an event is perceived as a stressor, it may initiate a
variety of physiological responses as the individual struggles to
adapt to the new situation. Just as moderator variables may influence
a person's perception of stressors, they may influence biological
reactions as well. Internal adaptation responses may occur in the
nervous, endocrine, and immune systems. Awareness of internal
responses may, in turn, affect perception in a feedback loop in a
manner to exacerbate or diminish the experience of threat. The
resultant fluctuation may upset the delicate balance necessary for
herpes simplex virus to remain in its quiescent existence. The virus
could then reactivate and result in an episode of recurrent herpes.
In the following section, potential mechanisms by which perceived
stress could result in a herpes recurrence will be considered.
Proposed Physiological Mechanisms
The precise triggering mechanism which leads to reactivation of
herpes simplex virus is elusive. While residing in the ganglion, the
virus neither destroys the latently infected cells nor is eliminated
by host defense mechanisms or prolonged antiviral chemotherapy.
Something occurs at the molecular level to stimulate it to migrate to
the skin surface and begin viral replication (Bierman, 1983; Stevens,
1975). To explain this phenomenon, two major hypotheses have been
proposed. In one hypothesis, the dynamic state hypothesis, a small
number of cells constantly replicate virus. Specific immune defense
responses involving antibody, lymphocytes, and interferon continually
work to localize and prevent the clinical manifestations of recurrent

37
infection (Stevens, 1975). This theory, also called the "skin trigger
theory," proposes that recurrences result when local defenses are
temporarily suppressed, or when slight physiological changes (other
than immune mechanisms) alter the balance in favor of the virus (Hill
& Blyth, 1976).
The static state hypothesis or "ganglion trigger" theory proposes
that the virus remains inactive in the ganglia until nervous system
stimulation provokes it to migrate toward the skin surface. In this
theory, the viral genome remains in an unproductive, quiescent state
(Stevens, 1975). At present, the weight of evidence favors the static
state hypothesis. Investigators that once believed in the dynamic
hypothesis (Hill but inactive in the ganglion. It is repressed and expresses few if
any viral antigens and, therefore, provokes no significant immune
response unless infection is reactivated (Blyth & Hill, 1984; Hill,
Blyth, Harbour, Berrie, & Tullo, 1983). The precise mechanism of
derepression allowing viral expression remains a mystery.
There is some indication that immune system status influences an
individual's vulnerability to recurrent herpes. This is particularly
true for cell-mediated; immune responses as opposed to humoral
immunity. Once infected with herpes simplex virus, individuals
develop neutralizing antibodies (humoral immunity) within 1 to 4 weeks
of primary infection (Nahmias & Roizman, 1973c). Unfortunately, the
antibodies are not effective in preventing reinfection or recurrent
infection in either healthy or immunocompromised individuals (Nahmias
& Roizman, 1973a; Rand, 1982; Rand, Rasmussen, Pollard, Arvin, &

38
Merigan, 1977). The herpes viruses, being capable of cell to cell
spread, are well adapted to evade antibodies. It has been
demonstrated that the titers of neutralizing antibody only
occasionally fluctuate before, during, or after recurrences (Nahmias &
Roizman, 1973b).
Fluctuations in cell-mediated immune responsivity, on the other
hand, appear to be related to recurrent herpes infections. Evidence
for this relationship comes from a variety of sources. A defect in
cellular immunity is proposed to account for the severity of herpes
infections in the newborn and the irnmunologically impaired host
(Nahmias & Roizman, 1973c). A quicker, more intense in vitro
lymphocyte transformation response was related to faster healing in
primary infection in one investigation. Thirteen of the patients were
followed for a period which included one or more recurrence. The
herpes simplex virus stimulation index was declining when recurrences
took place. Because the stimulation index was declining even among
patients who did not experience recurrence, authors suggest this
decline may be a necessary but insufficient correlate of recurrence
(Corey, Reeves, & Holmes, 1978). Similar results were obtained in an
investigation comparing the lymphocyte responsivity of patients with
frequently recurring oral herpes to a control group of seropositive
blood bank donors not reporting frequent herpes recurrences (Kirchner,
Schwenteck, Northoff, & Schopf, 1978). Responsivity of the two groups
was similar during infection, but during the disease-free interval
responsivity of the patient group was significantly lower. The
severity of infection in immunocompromised patients has been related

39
specifically to decreased responsivity of lymphocytes as well (Meyers,
Flournoy, & Thomas, 1980; Rand et al., 1977).
Another parameter of cellular immunity, the specific cytotoxicity
of peripheral blood mononuclear leukocytes (PBML) to cells infected
with herpes was found to fall to low levels during the week prior to
recurrent HSY in a group of 14 healthy seropositive adults, showing
enhanced reactivity during acute recurrent infection (Rola-Plezczynski
& Lieu, 1984). T-cells and the percentage of natural cytotoxic
lymphocytes were also found to decrease prior to recurrent HSV
episodes and to increase during recurrences. Fluctuations in
interferon, another component of cellular immunity, have likewise been
implicated in recurrences (Corey & Holmes, 1983; Rand et al., 1977)
and administration of interferon has significantly reduced the risk of
reactivation of patients undergoing surgery on the trigeminal ganglion
(Pazin et al., 1979).
To summarize, it appears that humoral immunity does not play a
major part in the manifestation or recovery from recurrent HSV
infection. This does not rule out the possibility that antibody may
contribute in combination with complement or as a mediator of
lymphocyte cytotoxicity. On the other hand, a relationship to HSV
infection status has been demonstrated for a number of cell-mediated
immune parameters including quantity and responsivity of T-
lymphocytes, interferon release, and natural lymphocyte cytotoxicity
specific for HSV.
A deficiency in immunologic competence, particularly impairment
of cell-mediated responsivity, could be a link between psychosocial

40
stress and recurrent herpes. Alterations as a result of stress have
been demonstrated in parameters of cellular immunity such as decreases
in lymphocyte responsiveness and natural killer cell activity
(Bartrop, Luckhurst, Lazarus, Kiloh, & Penny, 1977; Locke, 1982;
Kiecolt-Glaser etal., 1984). Changes in cel 1-mediated immune
functions are in turn related to herpes recurrences. An individual's
response to stress, therefore, could cause immunosuppression which
would allow viral replication. Physiological evidence for such a
connection will be reviewed as well as evidence for a more direct path
of influence which does not necessarily involve immune status.
There is consistent evidence that immunosuppression has an effect
on recurrences of herpes simplex virus. Seropositive transplant
patients whose immune responses are pharmacologically suppressed to
prevent organ rejection experience herpes lesions that are
inordinately severe, extensive, and long lasting. This occurs
sometimes, but not always, with increased incidence over a presurgery
recurrence base-rate (Blyth & Hill, 1984). For example, the
recurrence rate 5 to 15 weeks after bone marrow transplant is 60-80%
(Meyers et al., 1980). In the first 3 months following cardiac
transplantation, 83% of seropositive patients experienced herpes
recurrences of prolonged duration averaging over 2 months in duration
(Rand et al., 1977). Fifty percent of the seropositive renal
transplant paients developed herpes lesions of 2 to 16 weeks duration
within a 4-month follow-up period (Pass et al., 1979).
One of the early investigations to indirectly implicate stress as
a factor in the reactivation of HSV was done on rabbits. Laibson and

41
Kibrick (1966) experimentally induced recurrent ocular herpes in
latently infected rabbits by intramuscular administration of
epinephrine, a hormone well known to be released in excess during
periods of stress. Investigators demonstrated significant viral
isolation and infection in experimentally inoculated animals relative
to control animals. One could postulate an autonomic nervous system
link between stress and recurrence based on this evidence.
Reactivation of herpes in animals and humans has also been
accomplished by direct neural stimulation. Low levels of electric
current delivered via electrodes implanted over the trigeminal
ganglion of latently infected rabbits consistently induce viral
shedding and herpes recurrences, depending on the interval between
stimuli (Green, Rosborough, & Dunkel, 1981). In humans, trigeminal
surgery appears to exert a direct triggering effect. Between 83-94%
of seropositive individuals who undergo trigeminal surgery experience
recurrence and/or viral shedding within a week. These high incidence
recurrences are unremarkable in duration and severity and are
restricted to the area supplied by the stimulated or sectioned portion
of the trigeminal sensory ganglion (Carton & Kilbourne, 1952; Pazin et
al., 1979). Excessive; stimulation to peripheral tissue which receives
sensory nerves from the infected ganglion induces reactivation in
animals (Hill et al., 1983). Such stimulation could be analogous to
reactivation by sun, wind, or physical injury in naturally occurring
herpes episodes. One could extend the concept of reactivation by
direct neural stimulation to include excessive neural firing as might
result from psychological as well as physical stimulation. Thus, just

42
as hypnosis or expectancy (placebo effects) may act directly to alter
recurrence rates via a postulated central nervous system mechanism,
anxiety and stress may also act directly to increase recurrence rates
by causing excessive neural firing or by lowering the threshold for
neural firing.
It should be noted here that there appears to be a subtle
difference in how direct neural stimulation and immunosuppression
influence herpes recurrences. Direct neural stimulation results in
more immediate and higher rates of recurrence. These directly induced
recurrences are of average extent and duration. The recurrence rate
in immunosuppressed individuals is not necessarily higher, but the
recurrences are more extensive and long lasting. Of course, any
surgical treatment, whether immunosuppression is involved or not,
involves both physical and psychological trauma, providing a
confounded picture. Presumably, patients undergoing trigeminal or
transplant surgery experience grossly similar psychological
responses. A comparison of herpes simplex virus activation in these
two groups of patients suggests that direct neural stimulation is
associated with reliable reactivation of the virus, while
immunosuppression is more frequently associated with prolonged
severity and duration of recurrent herpes episodes.
One could conjecture more than one route by which stress could
influence the reactivity of HSY. Stress could directly stimulate the
latent virus via the central nervous system by causing increased
neural firing or a lower threshold for neural firing. Stress could
also be indirectly involved, by causing a deficiency in immune

43
surveillance. It is also possible that both are necessary to provoke
the clinical manifestation of herpes symptomatology, a trigger to
reactivate the virus and an immunocompromised environment in which it
may flourish. At the present time, the exact mechanisms that control
the latency and reactivation of herpes simplex virus in humans are
unknown. Immune response could play a vital role, but the
physiological state of the neuron could also be a key factor (Blyth &
Hill, 1984).
Because of its compatibility with existing evidence, the
supposition that both neural stimulation and immunosuppression are
necessary for herpes reactivation is intuitively appealing. However,
examinations of the validity of this and the other hypotheses present
serious difficulties. While it is known that some triggers are more
effective than others (e.g., trigeminal surgery more frequently
results in reactivation of HSV than sun exposure), the exact nature of
neural stimulation necessary to provoke a recurrence is unknown and,
therefore, difficult to measure or manipulate. Likewise, differences
in immunocompetency are difficult to define and even more difficult to
manipulate, particularly in humans, due to the ethical issues
involved. The preferred course at present is to continue research at
all levels of organization from the precise cellular/biochemical
measurements obtainable in the lab to the more global psychosocial
measurements obtainable through systematic naturalistic observations
of individuals over extended time periods. An integrated effort by
professionals with expertise in immunology, endocrinology, neurology,
pathology, and the behavioral sciences will be necessary to discern

44
meaningful patterns from divergent multiple measures. Ideally,
converging evidence from all of these disciplines will bring us closer
to the answers.
Present Investigation
The present study was undertaken to extend our knowledge of the
relationship between stress and health status. It was designed to
address methodological shortcomings of existing investigations.
Genital herpes was chosen as a disease model because of its recurrent
nature, the theoretical basis for a physiological mechanism linking
stress and recurrences, the compelling need for empirical information
regarding stress and recurrences, and the availability of an objective
laboratory diagnostic measure for disease confirmation. Prospective
in nature, the study design called for an evaluation of major and
minor life events upon enrollment and on a monthly basis throughout
the 6-month period of investigation. Documentation of herpes lesions
occurred within 24-36 hours of each recurrent episode. Thus, both
prospective and concurrent information regarding stress and
recurrences was collected. Additionally, three variables which have
shown promise in prior-' research as moderators of life stress were
assessed. The use of multiple regression techniques enabled the
clarification of the relative importance of social support, locus of
control, and arousal seeking tendency in influencing recurrence rates
in the face of stress. Thus the possibility of identifying subgroups
of individuals who were thought to be particularly vulnerable was
explored. In the present state of the art of psychobiological

45
research in which the single-cause, single-effect model is
acknowledged as simplistic, we must depend upon converging evidence
from the different levels of many disciplines to obtain a reasonable
picture of the many internal and external events and dynamic
interactions that ultimately result in a disease outcome (Schwartz,
1982).
This particular investigation involves the second and fifth
levels as illustrated in Figure 1: the perception of events as
stressful or not and the occurrence of viral reactivation as
manifested in recurrent herpes episodes. No attempt was made to
objectively monitor environmental events in the lives of subjects by
trained neutral observers. Nor have attempts been made to measure
neural, endocrine, or immune parameters. The independent variables in
this study consist of self-report measures of major and minor
stressors and social and personality variables which hypothetically
moderate the perception and influence of the stressor variables. The
dependent variable measured was a manifestation of genital herpes
symptomatology, the isolation of virus from a lesion. While this
particular investigation has not attempted to elucidate the complex
web of physiological events between stressor and symptomatology, it
has endeavored to shed light on the psychological conditions under
which a relationship between stress and illness exists.
Hypotheses
Several specific hypotheses were investigated in the present
investigation. As stated earlier, the present investigation was

46
designed to explore the relationship between stressful events and
recurrences using both prospectively and concurrently collected
information. The first hypothesis, using a prospective approach,
concerns the strength of the relationship between an individual's
experience of major life events and subsequent experience of herpes
recurrences. Specifically, the first hypothesis states:
1. The occurrence of undesirable life events prior to entry into
the study will be positively related to herpes recurrences during the
6-month period of investigation.
The next two hypotheses concern the concurrent experience of both
major and minor life events and herpes recurrences. Specifically,
2. The occurrence of undesirable life events during the 6-month
period of investigation will be positively related to herpes
recurrences over the same period of time.
3. The occurrence of monthly hassles will be positively related
to herpes recurrences during the period of investigation.
Additionally, it was hypothesized that three moderator variables,
social support, locus of control, and arousal seeking tendency, would
interact with each other in such a way that an individual's
vulnerability to herpes recurrences under high stress conditions would
be predictable on the basis of scores on the moderator variables.
That is,
4. The following variables which have previously been shown to
moderate the effects of life stress will be related to HSV recurrences
in the following ways: (a) Low levels of social support, a perception
of events as externally controlled, and a tendency to avoid arousing

47
situations will be associated with more frequent HSV recurrences, (b)
The antithesis of the above will serve to protect individuals from
frequent HSV recurrences. That is, individuals with a high level of
social support, with an internal perception of control, who tend to be
arousal seeking will have fewer HSV recurrences, (c) The probable
frequency of recurrence for a given individual will be predictable on
the basis of scores on the moderator variables.

CHAPTER TWO
METHODS
Research Design
The longitudinal design of this study allowed for the
investigation of prospective and concurrent information regarding the
occurrence of major and minor stressful events and episodes of
recurrent genital herpes. Data concerning past stressful events
collected from initial interviews and questionnaires provided
information to relate to ensuing recurrences of genital herpes
prospectively documented over the 6-month period of the study.
Concurrent information on major and minor stressful events was
collected monthly during a subject's 6 months of participation to
relate to documented recurrences for the same period of time. This
design provided the opportunity to avoid distortion due to
retrospective bias in an investigation of the relationship of major
and minor stressful events and herpes recurrences.
The assessment of; subjects' sense of perceived control, tendency
toward arousal seeking, and social support provided an opportunity to
evaluate the hypothesized stress-moderating effects of these
variables. In addition to studying the effects of the presumed
moderator variables separately, multivariate statistical techniques
made it possible to evaluate the manner in which these variables
related in an additive or interactive manner to possibly enhance or
48

49
diminish the resistance of individuals to recurrent episodes of
genital herpes in the face of life stress.
Sample Selection
Subjects were recruited over a 16-month period (March, 1983-July,
1984) through advertisements and referrals. Advertisements were
placed in both the university and local newspaper and on bulletin
boards on campus and in the J. Hillis Miller Health Center. Sources
of referral included the University of Florida Student Health Center,
the clinics of Shands Teaching Hospital, and participating
volunteers. Subjects were telephone-screened regarding their
suitability, availability, and willingness to make a 6-month
commitment to the sudy. The research was described as an effort to
understand more about the psychological factors involved in the
recurrence of genital herpes in order to improve the counseling
techniques of health care providers.
After obtaining informed consent, 10 cc of blood was drawn from
each participant. Those with a positive antibody titer to herpes were
enrolled in the study. Those subjects with a negative titer were not
enrolled unless they had a positive culture from a genital site to
herpes simplex virus.
Each of the 122 subjects who completed the full 6 months of
filling out questionnaires and having genital herpes recurrences
cultured received 100 dollars at the conclusion of the study for
his/her participation.

50
Measures
Stress Measures
The occurrence of major and minor events in the lives of subjects
was assessed by self-report questionnaires. Subjects completed the
questionnaires upon enrollment and monthly during the study.
Life Experiences Survey (LES). This questionnaire was used to
evaluate the experience of subjects with major life events. The LES
is comprised of 57 life events, 3 of which are fill-ins and the last
10 of which are specifically relevant to a college population
(Sarason, Johnson, & Siegel, 1978). Subjects indicate the frequency
of occurrence of each of the events over a specified time period.
They also rate the impact of each event on a scale from -3 (extremely
negative) to +3 (extremely positive), a zero rating designating a
neutral impact. Means and standard deviations are available for the
47- and 57-item versions based on 345 college students. Since many
subjects were college students, the longer inventory was used.
However, group statistical analyses were based upon the 47 items which
all subjects completed.
Desirable and undesirable events are assessed separately on the
LES, an advantageous feature for this investigation. Negative events
have been found to be more predictive of illness in general (Hotaling
et al., 1978; Sheehan et al., 1979) and genital herpes in particular
(Taylor, 1978; Watson, 1983). This is consistent with the literature
on adjustment (Mueller et al., 1977; Vinokur & Selzer, 1975).
Although impact (subject weighted) scores were available, the measure
of interest in this investigation was the frequency of undesirable

51
events. Several studies providing data on both simple frequency
scores and weighted scores have shown them to be equally predictive
where group data are concerned (Monroe, 1982b; Skinner & Lei, 1980;
Zimmerman, 1983). Developers of the scale have provided information
regarding significant positive correlations with several different
measures of anxiety, noncomformity, discomfort, depression, locus of
control, and a nonsignificant correlation with the Marlowe Crowne
Social Desirability Scale to demonstrate validity. Additionally, the
scale has been found to discriminate between normal controls and
counseling clients (Sarason et al., 1978). This scale has been used
in two retrospective investigations of subjects with recurrent
herpes. In one, the scale discriminated between subjects experiencing
high and low recurrences (Taylor, 1978). In the other, significant
positive correlations were found between the frequency and impact of
undesirable life events of the past year and the frequency and
duration of herpes recurrences for the past 6 months (Watson, 1983).
Hassles Scale (HS). The HS focuses on the minor events of the
past month, the irritating, frustrating, and distressing demands
characterizing daily transactions with the environment (Kanner et al.,
1981). The 117 items which include such things as care for pet,
having to wait, inconsiderate smokers, too many meetings, problems
with children, and preparing meals are checked if perceived as a
hassle and rated from 1 to 3 according to the severity of the
hassle. A frequency score is obtained by counting the number of
hassles checked. An impact score is obtained by summing the hassle
ratings. Norms are based on 100 middle class subjects ranging in age

52
from 45 to 64 who completed the scale for 9 consecutive months.
Test-retest reliability over that period is reported as .79 for
frequency scores, .43 for impact scores. Significant positive
correlations have been found with a negative affect scale, a life
events measure, a symptom check list (Kanner et al., 1981), and a
general health measure (DeLongis et al., 1982).
Moderator Variables
Upon enrollment in and termination from the research 6 months
later, subjects completed inventories to assess enduring
characteristics proposed to moderate the effects of stress on
illness. These included measures of social support, of perception of
internal or external controllability of events, and of arousal seeking
tendency.
Locus of Control (LC). The LC is a 29-item forced-choice
inventory including six filler items which assesses the degree to
which individuals view environmental events as being under their
personal control (Rotter, 1966). Subjects scoring low on the LC
(internals) tend to perceive events as being controllable by their own
actions, while those scoring high (externals) tend to view events as
i
being influenced by factors other than themselves. It is a measure of
generalized expectancy, or belief, rather than preference. The
measure is widely used and extensive data are available regarding
internal consistency, test-retest reliability, and discriminant
validity. Correlations with social desirability and intelligence are
low. Means and standard deviations on the scale are provided for

53
numerous groups, including large groups of university students
(Rotter, 1966). A number of studies investigating psychological
adjustment and illness as related to life stress provide support for
the role of perceived control as a moderator of life stress. That is,
an external LC score is associated with instability and illness under
conditions of increased life stress (Johnson & Sarason, 1978; Manuck
et al., 1975; Toves et al., 1981).
Arousal Seeking (AS). Arousal seeking is defined as the tendency
of individuals to engage in or avoid situations which increase
stimulation to a personal optimal level. The AS scale was developed
specifically to assess individual differences in this trait (Mehrabian
& Russell, 1974). Subjects are instructed to indicate their degree of
agreement or disagreement (on a 9-point scale) to each of 40 items.
Means and standard deviations are reported for two large samples of
college students. Adequate internal consistency and reliability have
been demonstrated. The AS scale correlated positively with a measure
of extroversion and negatively with measures of anxiety.
(Correlations with anxiety are significant, but weak; it is not simply
an anxiety measure.) Low correlations have been found with social
desirability and succorance (Mehrabian & Russell, 1974). In an
investigation of the relationship between life stress and measures of
anxiety, depression, and hostility, a significant relationship was
found only for subjects low on the arousal seeking dimension.
Negative life change in subjects who were low in arousal seeking
resulted in greater anxiety and hostility. This suggests that the

54
dimension plays a moderating role on life stress (Johnson, Sarason, &
Siegel, 1979).
Social Support Questionnaire (SSQ). Social support may be
defined as the degree to which individuals have access to social
resources or relationships upon which they can rely, particularly in
times of need. A person's satisfaction with the quality of such
resources is an important dimension in addition to the number of
persons in a support system. The SSQ measures the size and
satisfaction rating of an individual's social network (Sarason,
Levine, Basham, & Sarason, 1981). It consists of 27 two-part items,
one part to indicate the number of support people available under
particular circumstances, the other to indicate the respondent's
satisfaction rating for available support. Norms exist for a sample
of 602 university students. Internal consistency and test-retest
reliability are high. Correlations with scales of maladjustment show
an inverse relationship, suggesting that people who have fewer and
less satisfying social supports are more likely to be anxious, more
labile emotionally, and more pessimistic than others. Subjects low on
the SSQ tend to have an external locus of control and be relatively
low in self-esteem. ¡
There have been several studies which provide evidence for social
support as a moderator of life stress (de Araujo et al., 1973;
Hotaling et al., 1978; Nuckolls et al., 1972; Schaefer et al.,
1981). Those low in social support evidenced more maladjustment and
ill health under circumstances of high life stress. Therefore, high
levels of social support may play a stress-buffering role. In the

55
present investigation the SSQ was modified to include a pair of
questions regarding to whom a subject had disclosed he/she had herpes
and level of satisfaction with support received.
Physiological Measures
Screening procedure: HSV antibody. Since enrollment in the
study was limited to those who were herpes simplex virus (HSV)
seropositive, subjects were screened in order to ensure that they had
had prior exposure to HSV. This work was performed in a virology
laboratory and was done by a microtiter complement fixation (CF)
method previously described (Rand, Kramer, & Johnson, 1982). No
attempt was made to discriminate between antibodies to herpes simplex
virus-1 (HSV-1) and herpes simplex virus-2 (HSV-2) because (a) sera
that is seropositive by CF is seropositive for both HSV-1 and HSV-2 by
an accepted discrimination technique (ELISA); and (b) although HSV-2
usually causes genital herpes, HSV-1 is also a frequent cause.
Dependent variable: Virus isolation and transport. All subjects
were instructed to call the Clinical Research Center (CRC) as soon as
they were aware of a recurrent genital herpes lesion. An appointment
was made for a culture to be obtained within 24-36 hours by trained
i
personnel in the CRC. The CRC personnel were available on a 24-hour
daily basis.
Cultures for HSV were obtained by swabbing the base of unroofed
vesicular or ulcerated lesions with a cotton swab. Swabs were
immersed in transport media [minimal essential medium (MEM)
supplemented with 2 mm glutamine, 5000 u/1 penicillin, 50 ug/ml

56
streptomycin, 100 ug/ml gentamicin, 2 ug/ml amphotericin B and 1%
gelatin] and held at 4°C until transported to the laboratory; previous
work in the laboratory (unpublished) and elsewhere has shown this
method does not result in any significant decrease in recovery of
HSV-2 (Bettoli, Brewer, Oxtoby, Azidi, & Guinan, 1982; Yaeger, Morris,
& Prober, 1977). Cultures were then either frozen at -70°C or
inoculated immediately into primary cultures of human foreskin
fibroblasts (Rand et al., 1982). All isolates were identified by
typical cytopathologic effect.
Because subjects would not always be able to come to the CRC in
time, they were given Virocult swabs (Medical Wire and Equipment Co.,
Cleveland, OH) upon enrollment and instructed in the procedure for
obtaining self-cultures. Pilot work indicated that patients self-
obtained cultures were positive as frequently as staff-obtained
cultures. Herpes simplex virus has been shown to be stable for at
least 48 hours when refrigerated under these conditions (Medical Wire
and Equipment Co., product information).
Procedure
Initial Contact ;
Subjects were seen singly at the CRC by the investigator or a
trained research assistant. Confidentiality was ensured, study
procedures were explained in detail, questions answered, and informed
consent obtained. Blood was drawn for screening and subjects were
instructed to return to the CRC to have herpes lesions cultured every
time they had a recurrence during the 6-month period of the study.

57
They were provided with a Virocult swab and trained to obtain a self¬
culture in the event that it was impossible to get to the CRC within
24-36 hours. A culture could be safely kept in the refrigerator for 2
to 3 days until transport to the clinic was possible. A sample copy
of the LES was provided with a calendar on which to record major life
events. This eased the filling out of questionnaires each month and
helped to avoid overlap or duplication of major life events from month
to month. Subjects were then given a packet of initial questionnaires
and instructed in the procedure for completion. The packet contained
the LES (for the past 6-month and 1-year periods), the HS, LC, AS, and
SSQ questionnaires. An appointment was made for the second contact in
1 to 2 weeks time when the blood screening results would be available.
Second Contact
Subjects returned with the completed questionnaires of the
initial packet. Their questions were answered, and they were given a
second opportunity to sign the consent form to signify continuing
commitment to the investigation at a time when blood screening results
were known and they were thoroughly familiar with all procedures and
questionnaires. Subjects were not considered fully enrolled until
I
their second signature was obtained. Completed questionnaires were
reviewed for proper completion and a structured interview was
conducted to obtain demographic information and the subjects' history
of herpes. (Please see Appendix A for a copy of the structured
interview.)

58
Continuing Contact
Continuing contact was by mail, phone, and clinic visits. In the
last week of each month, subjects received by mail a packet containing
precoded questionnaires (LES and HS) to be completed on the final day
of the month with respect to events of the month; self-addressed,
stamped envelopes for their return; a newsletter; and educational
materials regarding genital herpes to maintain motivation. Subjects
were requested to date the major events reported on the LES to avoid
overlap with preceding and following months. Phone contact was made
to subjects to clarify material on the questionnaires, or to prod for
their return as necessary. Subjects initiated phone contact at any
time to ask questions about the study, or to obtain information about
herpes. Clinic visits occurred at any time during the 6 months of
participation for the purpose of obtaining culture material from
recurrent lesions. Subjects received counseling for disease
management by the CRC staff or the investigators if appropriate during
clinic visits.
Final Contact
A debriefing interview was conducted with each subject as soon as
the questionnaires for the sixth month were completed. Payment was
made to subjects following this final interview.

CHAPTER THREE
RESULTS
Sample Characters sties
The original sample consisted of 153 individuals. Five were
excluded because they did not have HSV antibodies in the initial
screening and thus were not subject to recurrent HSV. Attrition
occurred for the following reasons: (a) moved, j^=8; (b) too busy,
n_=4; (c) lost interest, jr=2; (d) received experimental vaccine, n_=l;
(e) believed participation worsened condition, n_=l; (f) visual
handicap, jt_=1 ; and (g) lost to follow-up, n_=8. Considering only the
seropositive subjects, this represents a loss of 26, or 18%, of the
eligible subjects.
The final sample consisted of 122 subjects, 83 of whom were
female (68%) and 39 of whom were male (32%). Average age was 27.3
(SD=6.1) with a range of 18 to 55. Educational level was high, with
the average number of years of education being 14.3 (range=9-18).
Sixty-three (52%) of the participants were currently in college or
graduate school. The majority of the participants (94; 78%) were
single, although 83 (68%) reported having a regular sex partner.
Subjects reported having herpes an average of 3 years, 11 months
(range of 1 month to 18 years). For the year prior to the study,
46.7% reported four or more recurrences, 35.8% reported one to three
recurrences, and 17.5% reported none.
59

60
Undesirable Life Events
The measures of major and minor life events in this study each
produced two scores of potential relevance to the present
investigation, a simple frequency of negative events score and an
impact score reflecting subjects' individual weighting of these
events. Life events scores of importance included a 6-month recall of
major events occurring just prior to enrollment, and monthly reports
of life events which were summed to produce a 6-month concurrent total
of events experienced during the investigation. The HS was also
completed monthly and summed to provide a 6-month total hassles
index. (See Appendix C for subjects' scores on all variables.)
Consistent with previous investigations (Zimmerman, 1983),
Pearson product-moment correlations of frequency and impact scores
resulted in extremely high correlations. The correlation between the
6-month retrospective LES frequency and impact scores was .95 (£=120;
£=.0001), between LES concurrent frequency and impact .95 (£=124,
£=.0001), and between concurrent HS frequency and impact .97 (£=124;
£=.0001). Although it might be argued that the weighted measures may
be more sensitive to an individual's perception of the stressful ness
of major and minor events, given the redundancy in information
provided by these indices, they were not included in further group
analyses. Simple frequency counts of negative events were used for
both the LES and HS measures as they were seen as providing the most
objective and straightforward indices of major and minor life changes,
respectively.

61
Table 1 provides means and standard deviations obtained on stress
measures for the 6 months prior to the study (past) and for the 6
months during the study (concurrent).
Table 1
Means and Standard Deviations of Past and Concurrent Stress Measures
Time Period
Stress Measure
Past 6 months
n_ = 120
Concurrent 6 months3
n_ = 117
Frequency of Major Negative
4.43
11.90
Life Events
(3.11)
(8.94)
Impact of Major Negative
9.57
21.02
Life Events
(7.16)
(17.24)
Hassles Frequency
-
121.48
(89.07)
Note. Means appear above standard deviations which are in
parentheses.
a Concurrent measures were computed by summing monthly scores over the
6-month period of the investigation.
Despite the fact that subjects dated major events for monthly
administrations of the LES to avoid overlap from month to month, the
average frequency of major negative life events is higher for the
concurrent 6-month period than for the 6-month period prior to entry
into the study, as recalled at the time of enrollment. There may be
several possible reasons for this discrepancy. First, it is possible
that the actual number of events experienced over the two 6-month time

62
periods differed. Secondly, the discrepancy may be attributed to
memory loss for events that occurred over the more remote 6-month
period prior to entry into the study (Monroe, 1982a; Paykel, 1983).
Memory loss for the concurrent period of investigation would be less
severe as recall was required over a shorter time period of 1 month.
Repeated reporting of major negative events that tend to recur
could also have contributed to the higher concurrent scores. For
example, the eight most frequently reported negative life events in
the concurrent information include events that could occur more than
once in a 6-month period. Arranged in order from the most frequent
they include sexual difficulties, major change in sleeping habits,
changed work situation, major change in financial status, major change
in eating habits, more or less arguments with spouse, breaking up with
boyfriend or girlfriend, and a major change in social activities.
These items were among those most frequently endorsed on the 6-month
recall as well. Such events could occur more often than once every 6
months and be counted a maximum of six times per subject in the
concurrent information (which is based on a total of six monthly LES
scores) but would only have been counted once on the 6-month recall of
major events experienced prior to the study.
It is of interest to compare LES scores of the present sample
with those of other groups. Normative data are provided on the LES
impact ratings of 345 male and female students enrolled in
introductory psychology classes (Sarason etal., 1978). Although
impact scores were not used in analyses in the present study, they
were available and are used here for comparison purposes since

63
normative data have been provided for LES impact but not LES frequency
scores. The mean 6-month recall negative impact score of herpes
subjects was 8.57 (^D_ = 7.16) as contrasted to a mean score of 4.66
(SD = 4.36) for males or 5.64 (SD_ = 6.43) for females for a 12-month
recall administration of the LES. One would have expected the figures
to be reversed since the data on the normative sample are based on a
time period that is twice as long (12 as opposed to 6 months). The
difference is even more striking for concurrent information with an
average frequency of major negative life events of 21.02 (SD =
17.24). The latter, as already discussed, includes information on
frequently occurring and potentially repeated major events. Authors
of the original measure suggest that the LES score of the student
(normative) population may be low relative to subjects from the
general population (Sarason et al., 1978). It appears that the
present, more heterogeneous sample has perceived relatively more
stress than the normative group.
Scores on the HS are more directly comparable to normative data
as both are based on monthly administrations (Kanner et al., 1980).
The 100 normative subjects were older, ranging in age from 45 to 64 as
opposed to the age range in the present sample of 18 to 55. The
average monthly hassles frequency score of the normative group was
20.50 which is very similar to the average for the herpes sample of
20.25.

64
Presumed Moderator Variables
Information regarding the herpes sample as a whole on arousal
seeking tendency, locus of control, and social support is provided
here for comparison purposes.
Subjects' scores on arousal seeking tendency were approximately
normally distributed with a mean of 35.02 (n_ = 120). This is slightly
higher than the average of 32.0 reported for the normative sample of
536 subjects (Mehrabian & Russell, 1974). The standard deviations of
the two samples are similar--28.58 for the herpes subjects and 29.0
for the normative group.
Locus of Control (LC) scores were also approximately normally
distributed with a mean score of 10.07 (^l¿ = 3.87, n_ = 122). Average
scores in the literature range from 5.94 (SD_ = 3.36, n_ = 155) for
Peace Corps volunteers to 9.22 (SD_ = 3.88, = 303) for college
students (Rotter, 1966). Although the standard deviations are
similar, the average score of the herpes sample is slightly higher,
indicating that, as a group, the present subjects tend to be more
externally oriented than others.
When the average Social Support Satisfaction (SSS) score is
compared to normative data, it initially appears that the herpes
subjects' SSS scores may be slightly lower. The average SSS score for
herpes subject is 141.87 (SD = 13.92, n_ = 119); for the normative
group of 602 undergraduates, the mean was 145.26 (SD_not provided;
Sarason et al., 1981). However, the distribution of scores from the
herpes subjects ranged from 2 to 213 and was highly skewed with most
of the scores grouped together at the very satisfied end of the scale

65
(median = 156.5). Subjects who were dissatisfied with their perceived
social support were not well represented in the present sample which
limits the generalizability of findings with regard to social support.
As a group, the herpes subjects appear to be slightly higher on
arousal seeking tendency and social support satisfaction and more
externally oriented than the respective normative samples described in
the literature.
As expected, major and minor events were significantly related.
The Pearson product-moment correlation of concurrent LES and HS totals
was .68 (jt_= 122; £=.0001). The HS scores were retained for analyses of
hypotheses specifically concerning minor events.
Definition of Recurrence
True herpes recurrent lesions do not always produce positive
culture results in the laboratory due to a relatively short duration
of viral shedding. Corey and Holmes (1983) report obtaining a 42%
culture positive rate from recurrent lesions. If the definition of
the dependent measure was restricted to culture-positive lesions, as
many as 58% of true episodes of herpes would be missed. If the
definition included all reported lesions, the value of laboratory
documentation in this investigation would be lost.
The decision was made to define a recurrence for the purposes of
the present study as any reported recurrence for an individual subject
for whom at least one culture-positive result was obtained during the
course of the study. This compromise was adopted to maximize clinical
sensitivity and scientific credibility. The different recurrence

66
definitions demonstrated a high degree of association with each
other. Spearman correlation coefficients of the number of recorded
recurrences and the number of positive cultures with the number of
recurrences given one positive were .66 and .71, both statistically
significant at the .0001 level. Furthermore, correlations of each of
the recurrence measures with the stress measures were very similar to
each other. (Refer to Appendix D for correlation coefficients of the
stress and recurrence measures.) Therefore, subsequent analyses
include all reported recurrences for subjects for whom at least one
positive was documented. Hereafter, unless otherwise stated, the term
"recurrences" will refer to those meeting the requirements of this
definition. During the study, 95 of the 122 total subjects reported
258 herpes lesions. Of these, 195 were cultured (76%), and 104 of
those cultured were positive (53%). This positive rate compares
favorably to that reported earlier (Corey & Holmes, 1983). There were
31 subjects whose lesions remained unconfirmed by laboratory
diagnosis.
Using recurrences as defined above, the hypotheses of the present
investigation were tested. For ease of presentation, results
concerning the existence and extent of a direct relationship between
stress and recurrences will be introduced first. Linear correlation
coefficients were used to examine this association. Following that,
analyses pertaining to the conditions under which such a relationship
exists are presented. Included here are the results of multivariate
analyses designed to investigate the combined influence of the stress
measures and presumed moderator variables. Finally, findings of

67
within-subject analyses accomplished for the purpose of exploring the
nature of the obtained relationship will be offered.
Stress/Recurrence Relationship
The first three hypotheses were tested using Spearman
correlations of stress scores and the number of herpes recurrences
experienced during the period of investigation. As defined above, all
recurrences were counted for each subject who had at least one
documented by viral culture during the 6-month study. This
information was correlated with the frequency of undesirable events
reported in the 6 months prior to the time of investigation to test
the first hypothesis. This is the purest prospective investigation of
stress and herpes recurrences since all stress information was
collected prior to the monitoring of recurrences. The result, based
on 120 individuals, was nonsignificant (£=.040; £=.656).
Because this result was surprising in light of two previous
retrospective investigations (Taylor, 1978; Watson, 1983), the
analysis was repeated for the subset of subjects who experienced
recurrences during the study, eliminating all subjects with zero
recurrences. This produced a subsample of recurring subjects, more
similar to the previous samples studied. The result was also
nonsignificant (£=-.040; £=53). These results indicate a total lack
of association between recent undesirable life events and subsequent
herpes recurrences, providing no support for Hypothesis 1.
In contrast, a significant positive relationship was obtained
between monthly totals of the frequency of reported undesirable life

63
events, summed over the 6-month period of the investigation, and
recurrences experienced over the same period of time. The correlation
obtained was .193 (£=.037; £=117). This significant positive
correlation indicates that increasing levels of reported negative life
events were associated with increasing recurrence rates when
concurrently monitored. Thus, some support was obtained for the
second hypothesis that stress and recurrences are concurrently
related.
With regard to Hypothesis 3, concerning minor life events
experienced over the same period of time, no significant relationship
was demonstrated. The correlation of the total HS scores reported
monthly and recurrences was only .019 (£=117). It appears that minor
life events are not significantly associated with herpes recurrences,
at least when monthly hassles scores are accumulated over a 6-month
period and related to recurrences over the same time period.
Influence of Moderator Variables
The fourth hypothesis concerns finding the best possible
combination of variables for predicting recurrences and evaluating the
influence of presumed moderator variables. Because it was the
particular pattern or combination of these variables that was of
interest, multivariate methods of analysis were chosen. It would have
been possible to use an analysis of variance procedure with the data
collected in this investigation in order to examine the differences in
recurrence rates associated with high or low scores on the moderator
variables and high or low frequencies of stressful events. However,

69
multivariate regression analyses make better use of the continuous
nature of the data and can provide a more precise description of the
relationships among the variables (Linn, 1982). In linear models,
multivariate techniques allow for the assessment of the effect of each
variable while statistically holding other variables of interest
constant. They also allow for nonlinear models which can include
interaction terms and therefore make it possible to assess the
moderating effects of one variable on another (SAS User's Guide,
1985).
Because the findings of the concurrent data provided the only
evidence for a stress/recurrence relationship, analyses employing
these concurrent data are the focus of the remainder of this report.
One analysis included all subjects, the second included only subjects
who experienced at least one recurrence during the study. Details of
the methods and results of these analyses and the follow-up procedures
used to verify results are described below.
When all individuals were considered in the analysis, a highly
skewed distribution of recurrences resulted due to the large number of
individuals with zero recurrences. Therefore, the recurrence variable
was categorized into three groups according to the number of
recurrences: (a) nonrecurrers who experienced no documented
recurrences during the study, (b) recurrers who had one or two
recurrences, and (c) recurrers who had three or more recurrences.
Logistic regression analysis was chosen as the most appropriate
analysis for use with this ordered categorical data (Harrell, 1983).
A logistic regression can be used to estimate nonlinear relationships

70
and takes into account the ordered nature of the recurrence categories
(Afifi & Clark, 1984; Harrell, 1983).
A forward stepwise logistic regression procedure was performed to
identify which, if any, variables and/or interactions were related to
recurrence groups. Single independent variables available to enter
the model consisted of the total number of negative major life events
reported monthly and summed over the 6 months of the study (LESNT),
the total number of hassles reported monthly and summed over the 6
months of the study (HST), and the following three variables which
were measured at the beginning of the study: locus of control (LC),
satisfaction with social support (SSS), and arousal seeking tendency
(AS). Two-factor interactions, which included at least one stress
measure (HST x LC, HST x SSS, HST x AS, LESNT x LC, LESNT x SSS,
LESNT x AS, LESNT x HST), and three-factor interactions, which
included one stress measure (HST x LC x SSS, HST x LC x AS,
HST x AS x SSS, LESNT x LC x SSS, LESNT x LC x AS, LESNT x AS x SSS),
were all available to enter the model if important in predicting
recurrences. The level of significance required for a variable to
enter and remain in the model was .05. The only term selected as
significant in this manner was the LESNT x SSS interaction (jK.OS).1
1 The logistic regression procedure was repeated with two
variations: (a) Subjects were eliminated whose cultures were never
documented by positive culture (n=31). In most cases, these cultures
were "documented" in the sense that a CRC nurse had judged the lesion
worthy of culture. Therefore, some were true recurrences for which an
attempt to isolate the virus failed, (b) Sex was included as a
variable in order to statistically control for a significant sex
difference on responses to the SSQ (_t( 117) =3.73; £=.0007). Removing
the "probable recurrers" from the group of nonrecurrers and
controlling for sex differences did not alter the logistic regression
solution.

71
A follow-up logistic regression analysis was performed with
LESNT, SSS, and their interaction as the independent variables. This
resulted in a prediction equation relating stress to recurrences. The
nature of the LESNT x SSS interaction, however, suggests that the
slope of recurrences on LESNT increases with increasing social support
satisfaction, meaning that at higher levels of SSS, there is a
positive relationship between stress and recurrences. Only those
extremely low on SSS demonstrate a negative relationship between
stress and recurrences. A change in slope occurred at approximately
100. For SSS scores greater than 100, the relationship between LESNT
and recurrences was positive, but negative for values less than 100.
In the present sample, SSS scores range from 69 to 158 with a mean of
141.2 and standard deviation of 19.8. The distribution is highly
skewed with more scores at the high end of the scale. Since less than
5% of the subjects' SSS scores fell below the critical score of 100
virtually no support was obtained for a beneficial moderating effect
of social support satisfaction on life stress. For the vast majority
of subjects, high life stress was associated with more frequent
recurrences.
Since previous investigations have focused on recurring subjects,
a regression analysis was performed using the subsample of the 52
recurrers for whom information on all independent variables was also
complete. The number of recurrences for an individual served as the
dependent variable. A square root transformation was applied to
stabilize the variance. A stepwise regression technique was utilized
in which variables which contribute significantly to the model enter,

72
remaining only if they contribute importantly in relation to all other
variables in the model (SAS User's Guide, 1985). The same independent
variables were available to enter the model as in the initial logistic
regression. The forward selection technique with a p_ level of .10
required for variables to enter and remain in the equation kept only
one term: the LESNT x LC interaction. The amount of variance
accounted for by this interaction was only 2%.
A follow-up regression analysis which included only LC, LESNT,
and their interaction was performed. This resulted in a prediction
equation which indicated that for individuals with a more external
locus of control (high values of LC), greater life stress resulted in
more recurrences. The opposite relationship held true for more
internally oriented individuals. The point at which the slope on
LESNT changed was approximately 3.43. So, for values of LC greater
than 3.43, LESNT was positively related to recurrences; for values of
LC less than 3.43 LESNT was negatively related to recurrences. In
this particular sample, LC scores ranged from 2 to 19 with a mean
score of 10.07 (SD_ = 3.87) and the distribution of LC scores was near
normal. As previously discussed, the herpes subjects tend to be more
externally oriented than normative groups (Rotter, 1966).
Follow-up regression analyses were performed separately for three
groups of individuals based upon their LC scores: (a) individuals
with LC scores between 2 and 8 (LC1), (b) those with LC scores between
9 and 13 (LC2), and (c) those with LC scores between 14 and 19
(LC3). Mean LESNT scores for these groups were 12.18, 13.48, and
13.75, respectively. The relationship between major life stress and

73
recurrences was different for these groups, as is illustrated in
Figure 2.
As can be seen by examination of this figure, the regression
lines for predicting the number of recurrences from an individual's
LESNT score are quite different in slope for individuals with the
lowest LC scores. Low LC scores are associated with belief in
personal (internal) control of events while high scores are associated
with belief in external control. Individuals with the lowest scores
(LC1) demonstrate a slightly negative nonsignificant slope, while
higher LC scores (LC2 and LC3) demonstrate a positive slope. In the
LC2 and LC3 groups, the estimated relationship was approximately the
same for both groups but only the slope for the LC2 group was
significant (£=.02). This illustrates the finding that life stress
has an important association with the probability of recurrence.
Since both regression analyses confirmed a stress/recurrence
relationship, the question of causality was raised. Does life stress
result in an increase in recurrences or vice versa? While it is not
possible to actually address the issue of causality as variables were
not experimentally manipulated in the study, a closer look at the
temporal relationship and direction of the relationship between stress
and recurrences appeared warranted. The following section is devoted
to follow-up analyses designed to suggest the direction of the
relationship between LES and recurrences.

Predicted Number of Recurrences
74
Number of Major Negative Life Events
Monthly Totals Summed Over Six Month Study
Figure 2
Predicted recurrences as a function of major negative life
events for three ranges of Locus of Control (LC) scores.

75
The Nature of the Life Stress/Recurrence Relationship
The method of data collection in this investigation allowed for a
more refined, within-subject analysis of the stress/recurrence
relationship. Information regarding life events was obtained at
regular monthly intervals, with recurrences being recorded by the date
of onset. This information was utilized in an attempt to clarify the
directionality of the stress/recurrence relationship.
Specifically, the hypothesis was that stressful events would
precede recurrences and this would be reflected by an increased number
of stressful events in the month prior to that of a recurrence. If on
the other hand, recurrences resulted in increased stress, this would
be reflected in an increased number of stressful events in the month
following the recurrence. To investigate this hypothesis, stress
difference scores were computed and summed for subjects who reported
recurrences. The difference scores were derived by subtracting the
stress score for the month following the month of a recurrence from
the stress score reported the month prior to a recurrence. If the
mean of these individual difference scores was significantly different
from zero in a positive direction, it would indicate that stress is
more likely to precede; than follow a recurrence and strengthen the
nature of the causal inference to be made. A nonsignificant or
negative result would weaken the case that stress causes recurrences,
perhaps suggesting that recurrent episodes of herpes result in
increased stress or that some third factor is related to stress and
recurrences causing similar changes to both.

76
Since it is possible that stress scores during months of a
recurrence could be inflated due to factors related to the recurrence,
these months were avoided in the analysis. That is, scores used were
from the first clear month (month without a recurrence) prior to a
recurrence and the first clear month following. Thus, if a person had
continuous recurrences in months 2, 3, and 4, the difference score was
computed between months 1 and 5. Difference scores on the LES could
not be obtained for recurrences falling in the first month since a
prior monthly stress score was not available. Similarly, recurrences
falling in the last month lacked stress information from a clear month
following the recurrence. In the case of more than one difference
score per subject, the scores were averaged for that subject. In the
event of recurrences separated by only 1 month, the month between
could be considered both a month prior and a month following.
Therefore, the month prior to the first recurrence and the month
following the last recurrence were used to compute difference
scores. Multiple recurrences within 1 month were treated as one
recurrence for the month.
The application of this procedure to the frequency of major life
events reported on the, LES on a monthly basis resulted in an average
difference score across subjects of 0.073. A one-sample _t-test for
this subsample of 55 subjects resulted in a jt-value of 0.28 which is
not significant (¿=.78). Because this was a within-subjects analysis
and weighted scores are more sensitive than simple frequency scores to
an individuals' perception of the stressfulness of events, this
analysis was repeated using impact scores. A similar nonsignificant

77
difference score was obtained. The average impact difference between
the month preceding and the month following a recurrence was 0.255
(_t=0.48; £=.63). These results together indicate that the frequency
and impact rating of stressful events preceding a recurrence are not
significantly different from the frequency and impact ratings of
stressful events following a recurrence. This does not resolve the
issue of directionality.
Because some of the negative difference scores were large and
could have masked the fact that stress more often precedes
recurrences, another test was performed. A simple count of negative
and positive difference scores was obtained and tested for
significance with the McNemar test for changes (Siegel, 1956).
Twenty-one subjects reported more stress preceding a recurrence, 17
reported more stress following. The rest of the subjects had equal
prerecurrence and postrecurrence scores. McNemar's test with a
correction for continuity resulted in a Chi-square of 0.237 which is
not significant at the .05 level. Thus, neither of the within-subject
analyses provided support for the notion that negative major life
stress is a significant factor in precipitating herpes recurrences.

CHAPTER FOUR
DISCUSSION
The major finding of this investigation was a significant
positive correlation between concurrent major life events and
documented recurrences of genital herpes. Subjects reporting more
stressful events over the 6-month period of study experienced more
recurrences. The predicted prospective relationship between stress
and recurrences was not confirmed by the data. Neither information
from the total sample nor from the subset of recurring subjects showed
any relationship between stressful events prior to the study and the
subsequent experience of herpes recurrences. Minor events, or
hassles, were not related to recurrences in any of the analyses.
The influence of several presumed moderators of life stress was
also investigated. Results differed depending on whether the total
sample or a subset of recurring subjects was considered. In the
former case, when information from all subjects was considered, an
interaction between social support and major negative life events was
found to be most predictive of recurrences. This interaction
indicated that high life stress was associated with more frequent
recurrences and that the effect of life stress became greater as
satisfaction with social support increased. This result was in direct
contradiction to prediction. In the latter case, when information
from only recurrers was analyzed, an interaction between major stress
78

79
and locus of control emerged as an important predictor of
recurrences. This finding was consistent with expectation. It
indicated that major stresses were associated with concurrent
recurrences for most subjects, excluding only those at the extreme
internal end of the locus of control distribution. Thus, an external
locus of control orientation and the experience of major stressors was
associated with higher recurrence rates. Arousal seeking tendency
failed to demonstrate any significant relationship to recurrence rate
in either analysis. In summary, with respect to those moderator
variables under study, only locus of control orientation interacted
with stress as predicted. Social support satisfaction had a
moderating influence directly opposite to that expected, and arousal
seeking tendency emerged as unimportant when other variables were
consi dered.
In the context of the present stress/illness literature, the
finding of a significant association between undesirable major life
events and an illness outcome measure is not new. It is important,
however, to have replicated this finding in an investigation designed
to avoid methodological flaws of earlier studies. In the present
study, stress elevations cannot be attributed to recall bias or a need
to justify illness (Rabkin & Streuning, 1976). Monthly reporting of
stressors was designed to avoid distortion due to forgetting (Funch &
Marshall, 1984; Monroe, 1982a; Paykel, 1983). Episodes of herpes,
objectively documented by viral culture, cannot be attributed to a
need to seek attention in the face of stress (Mechanic, 1975; Minter &
Kimball, 1978) or a lowered tolerance for discomfort due to stress

80
(Harney & Brigham, 1984). Therefore, the finding of a significant
association, albeit weak, is nonetheless important. The present study
employed stringent control measures and confirmed the findings of
previous studies. This enhances the credibility of earlier findings
and strengthens confidence in the existence of a stress/illness
association.
With regard to genital herpes recurrences in particular, the
finding is even more important. As reviewed earlier, the belief in a
stress/recurrence link is based entirely on anecdotal reports (Adams
et al., 1980; Blank & Brody, 1950; Janicki, 1971; tlllman, 1947),
survey research (Bierman, 1982; "Help membership HSV survey," 1981),
and retrospective investigations (Taylor, 1973; Watson, 1983).
Prospective investigations have either not found a significant
association, possibly due to inadequate methodology (Katcher et al.,
1973) or failed to systematically monitor stress (Freidman et al.,
1977; Luborsky et al., 1976). The best empirical evidence for a
stress/recurrence association has been provided by Taylor (1978) and
Watson (1983) in retrospective studies based on self-reported,
undocumented episodes of recurrent herpes. Taylor (1978) found that
women who reported a h,igh rate of recurrence also reported a higher
frequency of stressful events over the previous year. Watson's (1983)
replication of the finding in a larger sample of males and females
strengthens confidence in it. The correlation between major negative
life events and recurrences found in the present study (r_=.19) is in
agreement with (significant and positive), but weaker than, those

31
found by Watson (1983; r=.32, r=.41). However, it is based on
evidence less subject to confounds than prior research.
The degree of confidence one may have in the findings of a
particular investigation can be related to the type of evidence it has
provided. One can conceptualize at least three levels of evidence
related to methodology in stress/illness research. Prospective
research in which stress is evaluated prior to the occurrence of
illness provides the purest and therefore the strongest evidence.
Concurrent research in which stress and illness are evaluated over the
same time period is somewhat weaker. The significant findings of the
present investigation are based on concurrent stress and recurrence
measures. Retrospective findings which form the bulk of evidence are
the weakest because the possibilities for distortion are highest.
Retrospective investigations are valued more for their heuristic
quality than for definitive findings.
The ability to demonstrate a prospective relationship to a large
extent depends upon the timing of event recording and occurrence of
illness. The impact that life events have may vary over different
periods of time (Sarason et al., 1982). Whether remote or recent
events affect a particular illness is also dependent upon the
characteristies of the illness. The timing of event-reporting in
life/stress research has been generally a matter of convenience or
convention. In an early investigation by Rahe and Holmes (cited in
Holmes & Masuda, 1974), physicians retrospectively reported stressful
events and health changes over a 10-year period. More recently, in
light of information regarding forgetting and distortion (Funch &

82
Marshall, 1984; Monroe, 1982a), shorter intervals have been used. A
decision regarding the optimal reporting interval to use in order to
observe a relationship could also be made considering physiological
information of the particular disease in question and related
stress/illness research findings.
With respect to genital herpes, there are reasons to look both at
long and very short latency periods. A long latency period is
suggested by Watson's (1983) research. He found a strong relationship
between stress during one 6-month period and recurrences in the
following 6-month period. The present study failed to replicate those
findings prospectively. Stresses in the 6-month period prior to the
present study were not related to documented recurrences during the 6-
month study. To investigate the possibility that stressful events
which preceded the study were related to recurrences in the first few
months as opposed to the entire 6-month period of the study, Spearman
correlations were done between LES scores for the 6-month period prior
to the study and recurrences in the first 2 months and the first 3
months of the study. These were -.13 (£=.23, £=85) and -.07 (£=.48,
£=85), respectively, indicating no significant relationship between
stress recalled for the 6 months prior to the study and recurrences in
the first 2 or 3 months of investigation. It is possible that the
earlier finding may have been influenced by retrospective bias or the
homogeneity of the sample (Watson, 1983) and resulted in a spurious
finding. More research is needed to resolve this issue.
An argument can be made on the basis of the present investigation
for a shorter latency. The significant findings were in the

83
concurrent information as opposed to the prospective information. The
attempt to use the monthly nature of data collection to examine the
essence of this stress/recurrence relationship more closely did not,
however, suggest elevated stress in the months prior to recurrences.
Although there could be many explanations for this observation, one
strong possibility is that the temporal relationship between stress
and herpes is of such brief duration that monthly measures are too
infrequent to capture the variability in stress that relates to
recurrences. If stress is causally related to recurrences and the
latency between the occurrence of a stressful event and a herpes
recurrence is a matter of 5 days or less, an event that occurs in the
beginning of the month could be related to a recurrence in the same
month. The elevation of stress that occurred prior to the recurrence
would not have been reflected in the score of the preceding month.
Similarly, if recurrences are causally related to stress and a
recurrence occurs in the beginning of the month, the resultant stress
could be reflected in LES scores in the same month and may not be
reflected in the scores of the following month. Thus, the analysis
based on monthly measures may not have been sensitive enough to detect
stress elevations that may be relevant regarding the directionality of
influence between stress and herpes.
The latency between the occurrence of stress and reactivation of
HSY may, in fact, be only a matter of days or hours. In support of
this, a latency of 1 to 2 days was reported by Herlig and Hoff (cited
in Janicki, 1971) and Ullman (1947) between emotional upset and
appearance of blisters. Direct neural stimulation such as occurs

84
during trigeminal surgery reliably results in recurrent herpes
episodes in 3-5 days in humans (Carton & Kilbourne, 1952; Pazin et
al., 1979). In an animal model, electrical stimulation of the nerve
produces herpes blisters within 3 days (Green et al., 1981).
The temporal relationship is an important consideration in the
design of life/stress studies. The present findings suggest the
possibility of a brief latency between stress and herpes. Therefore,
a recommendation for future research is to evaluate stress on a weekly
or even daily basis, while continuing to evaluate the possibility of
longer latencies. The data from more frequent monitoring can be
collapsed as desired to evaluate the impact of stress over a variety
of time intervals.
One of the aims of the current research was to enhance the
ability to predict illness outcome by evaluating minor stressors
(hassles) and presumed stress moderators as well as major life
events. There is a notion that the ill effects of major life stress
may be mediated by an increase in the number of hassles experienced
(Kanner et al., 1981). That is, the ill effects of major events
(e.g., moving) may be due to increased hassles (e.g., trouble getting
meals, being lonely, too many things to do). In the present research,
hassles did not demonstrate a direct relationship to recurrences and
the inclusion of hassles did not enhance the predictability of
recurrences in either regression analysis. In this sample at least,
the association of major negative life events and recurrences was
independent of an association with hassles.

85
With regard to the moderator variables, it was predicted that
they would interact with each other in such a way that a person's
scores on each of the measures would help to determine susceptibility
to recurrences under varying conditions of stress. This did not occur
in accordance with expectations for each measure. Of the three
presumed moderators chosen for study, only locus of control
orientation interacted with major stress as predicted. It was
hypothesized that individuals who do not feel in control of the events
in their lives would react more to stress and therefore be more likely
to have recurrences if stressed. There is support for this with
regard to adjustment outcomes (Johnson & Sarason, 1978), treatment
seeking (Manuck et al., 1975), and retrospectively reported herpes
recurrences (Watson, 1983). In general, a relationship is seen
between stress and illness only for externally oriented individuals.
The present study found support for this in the subsample of subjects
who had recurrences. No relationship was found between stress and
recurrences for individuals on the extreme low end of the scale
(internals). A significant positive relationship was found between
stress and recurrences for externals providing further support for
locus of control as a Stress moderator.
Social support has been widely researched by investigators using
a wide variety of measures. In general, either an overall beneficial
effect or a buffering role in relation to life stress has been
attributed to social support (Cohen & Wills, 1985). These beneficial
effects have been more consistently demonstrated with psychological
outcomes than with illness measures however. Watson (1983) found a

86
direct effect on genital herpes recurrences such that those high on
social support (perceived helpfulness) had significantly fewer
recurrences. It was important to replicate these findings in a
prospective investigation because retrospective bias and recruitment
procedures in the earlier study limit the generalizability of the
findings. Forty of the 46 subjects were recruited through a herpes
self-help group which may have biased the findings regarding perceived
helpfulness (Watson, 1983).
The findings of the present study with regard to social support
are not in accord with those suggesting either a beneficial main
effect or those demonstrating a mitigating effect on stress. The
nature of the moderating role uncovered for social support revealed
instead a significant positive relationship between stress and herpes
for individuals high on social support satisfaction. That means
people who reported more satisfaction with social support also had
more recurrences when under stress. This unexpected finding was
closely examined. In the present investigation, social support scores
were highly skewed. Most subjects' scores suggested that subjects
were either very satisfied or very very satisfied with support
received. It is possible that individuals in this particular sample
were predominantly very active socially and happy with that. However,
subjects may have overextended themselves socially, thus negatively
affecting their resistance to recurrences. Unfortunately, the
restricted nature of the sample with regard to social support limits
the generalizability of these findings.

87
It was considered that subjects reporting high satisfaction with
social support ¡nay have been more sexually active which, in turn, led
to more frequent recurrences of genital herpes. Because information
was available on the frequency of sexual activity on a monthly basis,
this possibility was empirically examined. Frequency of sexual
activity was found to be related to social support satisfaction
(£=.21; £=.024) but the number of recurrences was negatively related
to frequency of sexual activity (£=-.25; £=.007). This suggests that
the number of recurrences may influence the frequency of sexual
activity rather than vice versa.
It is unfortunate that the sources of support are not
identifiable on the SSQ used in this investigation. It was possible
to discern network size and perceived quality of support, but not
possible to know if the perceived sources of support were family,
friends, professionals, peers, confidante, or a significant other.
With regard to genital herpes in particular, the support of a
significant other could be more important than others.
It was also not possible to differentiate between types of
support on the social support measure. Cohen and Wills (1985) provide
a useful typology of four support resources: esteem support,
informational support, social companionship, and instrumental
support. They suggest that there must be a reasonable match between
coping requirements and available support in order for stress
buffering to occur. Relating this to the present findings, it is
conceivable that respondents' reported perceived satisfaction on the
SSQ is more reflective of the type of support labeled social

88
companionship by Cohen and Wills (1985). Thus, despite the tendency
to be quite pleased with available support, it may not be the type of
support that assists one in dealing with increased stress. As
suggested by Graves and Graves (1985) whose measures of illness
symptoms also provided evidence contrary to a stress-buffering role
for social support, perhaps the obligatory nature of dense social
networks is maladaptive under conditions of life change.
Together with the wide range of findings regarding social support
in the literature (see reviews by Cohen & Wills, 1985; Thoits, 1982),
the present results further emphasize the complexity of the phenomenon
of social support, and the need for more specific information
concerning source and type of support.
Arousal seeking tendency was hypothesized to exert a stress
moderating role on the basis of studies dealing with adjustment
outcomes. Subjects low on arousal seeking tendency have been shown to
be more subject to psychological discomfort under high stress
conditions than high arousal seekers (Johnson et al., 1979; Smith et
al., 1973). The present investigation unsuccessfully attempted to
replicate and extend those findings with regard to a physical outcome
(genital herpes recurrences).
Although the logistic regression and stepwise regression
procedures resulted in interesting findings, only 3% of the variance
in recurrences was accounted for. Furthermore, because a different
moderator variable was chosen in each regression analysis depending
upon whether the full sample or only recurrers were utilized, it
suggests the effects may not be robust. Social support was important

89
only when predicting who will or will not have recurrences, locus of
control was important when predicting how many recurrences the
recurrers would have.
Thus, the present investigation has provided support for a
significant but weak association between stress and illness. This
raises questions of how to proceed in future investigations of the
nature of recurrences and their relationship to stress. In order to
enhance predictability as originally planned, further methodological
refinements will have to be accomplished. The use of more frequent
stress monitoring to evaluate a variety of impact intervals and a more
specific measure of social support have already been recommended.
On the basis of the suggestive results of the present study
concerning the effects of social support and locus of control,
continued exploration of individual differences appears warranted.
Other psychological factors might also be considered. While it was
inferred that the three characteristics chosen for study in the
present investigation would enhance one's ability to cope with stress
and thus mitigate the illness-producing effects of stressful life
events, coping skills such as problem-solving ability could be more
directly evaluated. ¡
In addition to psychological factors, the role of biological
factors should be examined. Results of an earlier prospective study
of oral herpes suggest that biological factors are better predictors
of oral herpes recurrence than psychological factors including mood
and social assets. Disease history and illnesses accounted for 80-902
of explained illness (Friedman et al., 1977). A measure of illness

90
history and predisposition could potentially improve predictability.
Because of the interest in the mechanism of recurrence, it might also
be useful to evaluate hormonal and immunological parameters to examine
fluctuations which relate to stress and recurrence onset. To maximize
the probability of observing recurrences, high recurrers could be
evaluated in a single-subject design with frequent stress and
physiological monitoring.
The present findings regarding major negative life events and
recurrent herpes provide limited empirical support for the long held
notion of a stress/recurrence link. Because this evidence is
consistent within the study and based on concurrent findings, it is
stronger than that of previous retrospective findings. However,
despite stringent control measures, 96-97% of the variance with regard
to genital herpes recurrences remains unexplained. It behooves
researchers and clinicians to collaborate with professionals in an
interdisciplinary manner to further untangle the complexities of
recurrences.

APPENDIX A
STRUCTURED INTERVIEW
Initial Structured Interview
Demographic Information
1. Sex:
(1) F
(2) M
2. Age:
3. Marital Status:
(1) Single
(2) Living with significant other
(3) Married
(4) Separated
(5) Divorced
(6) Widowed
4. How many children do you have?
(1) 0
(2) 1
(3) 2
(4) 3
(5) 4
(6) 5 or more
5. What level of education have you completed?
(1) Grade school
(2) High school graduate
(3) Some college
(4) College graduate
(5) Some graduate work
(6) Master's degree
(7) Ph.D., M.D., or other advanced degree
6. Are you presently a student?
(1) Yes
(2) No
91

92
7. Have you had any major illnesses over the past 6 months?
(1) Yes
(2) No
a. Do you have any major illnesses or chronic conditions?
b. Are you taking any medication?
8. What prompted you to volunteer for this study?
9. When did you first experience problems with genital herpes?
10. When was genital HSV first diagnosed?
By whom?
How?
11. When was your most recent HSV recurrence?
12. How many HSV recurrences have you had in the last year?
None 1-3 4 or more
13. What do you think causes recurrences for you?
14. Do you at present have a regular sex partner? Yes No
15. Does your partner have herpes? Yes No
With respect to the last 6 months, what has been your usual frequency
of:
16. Geni tal/geni tal contact?
17. Oral/genital contact?
18. Masturbation?
19. With approximately how many different partners have you had
sexual contact over the past 6 months?
20. Have you had sexual contact with any partners who have genital
HSV over the past 6 months?

93
21. Has there been a change in your usual pattern of sexual activity
over the past 6 months?
If so, describe:
To what do you attribute this change?
22. Do you have any information that you want to add to what we've
already talked about?

APPENDIX B
SUBJECTS' TOTAL SCORES ON STRESS, MODERATOR,
AND RECURRENCE VARIABLES
Key to variable names:
ID
Subject identification (changed for publication).
LESN-6
Frequency of major negative life events for 6 months
prior to study.
LESNT
Frequency of major negative life events reported monthly
and summed for 6 months of study.
LES IMP
Impact of major negative life events reported monthly and
summed for 6 months of study.
HAST
Frequency of hassles reported monthly and summed for 6
months of study.
HASIMPT
Impact of hassles reported monthly and summed over 6
months of study.
AS
Arousal seeking score.
SSS
Satisfaction with social support.
LC
Locus of control score.
HSVREC
Number of recurrences recorded.
HSVPOS
Number of; recurrences documented by positive culture.
NUMRECIP
Number of recurrences counted if one was documented
positive.
94

ID
1
2
3
4
5
6
7
8
9
10
1 1
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
95
LESN-6
LESNT
LESIMP
HAST
HASIMPT
AS
SSS
LC
HSVREC
HSVPOS
NUMRECIP
6
18
21
67
75
49
162
9
12
9
12
7
13
14
109
139
35
128
12
0
0
0
12
38
82
166
228
5
156
10
4
4
4
3
1 1
20
120
202
49
153
16
0
0
0
6
7
17
40
69
.
162
2
5
0
0
2
2
3
29
41
4
157
10
0
0
0
1
7
11
82
133
-23
133
5
3
1
3
7
15
27
228
316
.
162
11
0
0
0
5
13
23
46
53
76
151
4
3
0
0
2
8
23
76
136
66
139
5
0
0
0
0
0
0
16
16
42
157
3
2
0
0
3
5
8
74
86
-7
164
8
5
3
5
3
20
40
153
277
64
125
16
2
1
2
3
.
99
136
23
160
1 1
4
1
4
3
13
22
149
225
-42
147
7
7
0
0
4
24
49
92
153
25
157
7
8
1
8
1
2
3
33
40
51
148
8
1
0
0
2
10
19
100
126
50
117
9
1
0
0
3
8
14
50
69
30
142
1 1
3
2
3
6
24
55
192
415
-8
103
16
10
3
10
6
10
15
56
63
34
115
6
1
0
0
15
20
47
191
413
68
163
8
1
1
1
0
3
3
33
34
94
143
3
0
0
0
1
3
3
27
37
24
86
7
5
2
5
10
12
29
220
398
43
147
9
1
1
1
6
2
3
79
88
16
167
6
2
1
2
5
.
.
.
.
79
162
8
2
1
2
7
32
73
253
485
25
142
17
0
0
0
5
18
33
115
190
54
126
15
3
0
0
7
1 1
22
90
138
50
149
10
3
1
3
4
9
14
59
86
42
126
1 1
3
0
0
1
12
23
82
107
63
142
15
0
0
0
7
5
7
40
52
68
138
5
0
0
0
2
3
4
97
153
13
142
1 1
0
0
0
6
29
67
214
47 1
80
158
12
2
0
0
6
5
10
48
61
78
138
10
3
3
3
1
6
13
91
173
-2
110
1 1
0
0
0
2
16
27
69
95
39
160
7
0
0
0
7
12
24
62
70
39
160
9
0
0
0
3
5
8
40
45
-16
139
1 1
1
1
1
5
8
19
1 1 1
120
39
156
10
2
0
0
0
9
10
99
128
38
131
7
5
4
5
4
5
6
45
54
62
140
13
1
1
1
0
0
0
53
59
40
165
5
3
0
0
.
.
•
258
401
17
99
14
0
0
0
5
8
9
65
89
32
161
8
5
1
5
1
5
10
47
66
17
129
13
1
l
1
4
12
24
103
211
22
162
12
0
0
0
5
9
15
40
63
19
138
14
3
0
0
6
12
29
78
113
32
152
1 1
4
1
4

ID
LESN-6
LESNT
LESIMP
HAST
HASIMPT
51
3
14
30
62
81
52
6
11
16
128
165
53
3
8
1 1
80
91
54
6
27
68
332
729
55
3
17
30
125
179
56
5
10
18
140
195
57
4
1 1
18
1 17
163
58
14
19
37
212
318
59
5
9
12
87
153
60
8
10
12
100
114
61
3
5
5
36
37
62
13
17
29
157
278
63
3
17
31
118
192
64
1
1
2
59
76
65
3
10
14
88
97
66
2
1
2
45
48
67
1
7
13
150
239
68
6
16
24
37
58
69
4
13
16
77
101
70
1
4
6
82
96
71
0
4
7
23
27
72
8
45
68
378
656
73
4
8
13
91
99
74
2
22
26
.
75
4
15
30
106
138
76
1
15
35
193
301
77
7
25
43
96
128
78
3
8
14
109
185
79
4
11
15
92
105
80
5
20
29
166
209
81
3
8
16
57
88
82
3
2
4
49
67
83
3
4
5
217
243
84
7
16
22
148
214
85
1
1 1
23
155
249
86
6
1 1
20
69
164
87
8
21
34
151
194
88
6
27
51
87
117
89
5
16
27
59
70
90
7
28
43
230
378
91
1
5
6
.
#
92
0
0
0
45
53
93
2
0
0
73
90
94
6
18
30
145
179
95
2
6
1 1
250
339
96
5
7
13
129
243
97
1
6
16
218
396
98
5
.
.
.
.
99
4
6
1 1
55
75
100
7
18
25
82
150
sss
LC
HSVREC
HSVPOS
NUMRECIP
150
9
1
0
0
155
16
0
0
0
125
9
1
0
0
.
17
0
0
0
148
12
3
0
0
124
13
0
0
0
109
14
1
l
1
125
10
2
1
2
1 44
5
1 4
5
14
121
8
0
0
0
141
15
1
1
1
120
12
0
0
0
134
9
5
3
5
120
13
0
0
0
166
14
0
0
0
140
9
0
0
0
138
6
1
0
0
151
6
1
1
1
138
1 1
l
0
0
168
12
4
4
4
106
1 1
1
1
1
141
12
9
4
9
158
7
0
0
0
154
9
4
3
4
165
15
3
2
3
146
10
1 1
5
1 1
1 42
19
1
1
1
.
19
3
1
3
137
10
2
2
2
152
15
1
1
1
123
15
2
0
0
159
14
1
1
1
134
9
0
0
0
134
14
2
0
0
137
10
5
3
5
160
1 1
2
1
2
104
16
0
0
0
135
9
2
1
2
119
12
1
1
1
143
6
4
1
4
154
10
4
0
0
139
12
0
0
0
119
3
5
1
5
150
4
2
0
0
128
8
1
0
0
140
9
1
0
0
101
16
0
0
0
115
2
1
0
0
160
1 1
1
1
1
162
6
2
1
2
• AS
15
34
14
77
10
47
6
-12
-7
12
93
48
29
-1 1
67
32
91
67
22
45
22
12
13
9
26
64
-9
54
37
36
44
59
30
49
44
40
62
42
20
25
31
71
45
41
8
28
-28
79
28
30

ID
LESN-6
LESNT
LESIMP
HAST
HASIMPT
AS
SSS
LC
HSVREC
HSVPOS
NUMRECIP
101
4
25
30
139
182
64
123
13
1
0
0
102
12
23
41
453
644
32
106
12
2
1
2
103
.
49
82
630
1096
36
130
13
1
1
1
104
5
10
14
237
384
97
168
9
4
0
0
105
5
6
9
146
184
24
152
18
1
0
0
106
7
15
26
162
233
42
154
10
1
0
0
107
3
7
12
79
80
1 1
167
17
5
2
5
108
12
3
6
130
206
-20
166
6
1
0
0
109
1 1
24
39
207
333
85
1 49
14
3
2
3
110
3
10
21
83
143
28
149
6
0
0
0
1 1 1
2
4
6
68
98
65
153
4
0
0
0
112
5
21
27
248
341
33
151
6
4
3
4
113
1
0
0
74
99
10
167
1 1
2
0
0
114
3
3
6
94
116
60
99
8
0
0
0
115
4
18
32
132
158
-9
127
5
0
0
0
116
5
7
9
98
125
85
168
6
0
0
0
1 17
4
10
15
194
285
1 1
156
10
1
0
0
118
1 1
«
.
.
.
25
.
9
0
0
0
1 19
3
10
17
201
286
50
168
7
0
0
0
120
1
7
1 1
233
255
-25
159
12
1
1
1
121
1
5
7
85
90
56
158
4
0
0
0
122
1
11
20
129
189
48
110
7
4
3
4

APPENDIX C
SPEARMAN CORRELATION COEFFICIENTS AND SIGNIFICANCE LEVELS
FOR REPORTED STRESS FREQUENCY AND RECURRENCE DEFINITIONS
Recurrence Definitions
Reported Stress Frequency
Major Negative
Life Events
Hassles
past 6 months
n_=120
concurrent
n=117
concurrent
ji=117
Total recorded
.04
.18*
-.00
(.657)
(.049)
(.991)
Documented by positive
.04
.21*
.02
culture
(.692)
(.021)
(.869)
At least one documented
.04
.21*
.02
(.656)
(.020)
(.857)
Note: Correlations appear
above significance
levels which
are in
parentheses.
* JL.S *°5
98

REFERENCES
Adam, E., Dressman, G. E., Kaufman, R. H., & Melnick, J. L. (1980).
Asymptomatic virus shedding after herpes genitalis. American
Journal of Obstetrics and Gynecology, 137, 827-830.
Afifi, A. A., & Clark, V. (1984). Computer aided multivariate
analysis. Belmont, CA: Wadsworth.
Bartrop, R. W., Luckhurst, E., Lazarus, L., Kiloh, L. G., & Penny, R.
(1977). Depressed lymphocyte function after bereavement.
Lancet, 1, 834-836.
Bedell, J. R., Giordani, B., Armour, J. L., Tavormina, J., & Boll, T.
(1977). Life stress and the psychological and medical
adjustment of chronically ill children. Journal of
Psychosomatic Research, 21, 237-242.
Berkman, L. F., & Syme, S. L. (1979). Social networks, host
resistance, and mortality: A nine-year follow-up study of
Alameda County residents. American Journal of Epidemiology,
109, 186-204.
Bettoli, E. J., Brewer, P. M., Oxtoby, M. J., Azidi, A. A., & Guinan,
M. E. (1982). The role of temperature and swab materials in
the recovery of herpes simplex virus from lesions. Journal
of Infectious Diseases, 145, 339.
Bierman, S. M. (1982). The myth and realities of genital herpes.
The Journal of Mind and Behavior, 3, 27-46.
Bierman, S. M. (1983). A possible psychoneuroimmunologic basis for
recurrent genital herpes simplex. Western Journal of
Medicine, 139, 547-552.
Bierman, S. M. (1985). Recurrent genital herpes simplex infection:
A trivial disorder. Archives of Dermatology, 121, 513-517.
Blank, H., & Brody, M. W. (1950). Recurrent herpes simplex: A
psychiatric and laboratory study. Psychosomatic medicine,
12, 254-260.
Blyth, W. A., & Hill, J. J. (1984). Establishment, maintenance, and
control of herpes simplex virus (HSV) latency. In B.T. Rouse
and C. Lopex (Eds.), Immunobiology of herpes simplex virus
infection (pp. 9-32). Boca Raton, FL: CRC Press.
99

100
Bradley, C. (1979). Life events and the control of diabetes
mellitus. Journal of Psychosomatic Research, 23, 159-162.
Bramwell, S. T., Masuda, M., Wagner, N. N., & Holmes, T. H. (1975).
Psychosocial factors in athletic injuries. Journal of Human
Stress, 1, 6-20.
Brown, G. W., Bhrolchain, M. N., & Harris, T. (1975). Social class
and psychiatric disturbances among women in an urban
population. Sociology, 9, 225-254.
Carton, C. A., & Kilbourne, E. D. (1952). Activation of latent
herpes simplex by trigeminal sensory-root section. New
England Journal of Medicine, 246, 172-176.
Cleary, P. D., & Angel, R. (1984). The analysis of relationships
involving dichotomous dependent variables. Journal of Health
and Social Behavior, 25, 334-348.
Cleary, P. J. (1980). A checklist for life event research. Journal
of Psychosomatic Research, 24, 199-207.
Cohen, L. H. (1982). Life changes and the sensation seeking motive.
Personality and Individual Differences, 3, 221-222.
Cohen, S., & Hoberman, H. (1983). Positive events and social
supports as buffers of life change stress. Journal of
Applied Social Psychology, 13, 99-125.
Cohen, S., & Wills, T. A. (1985). Stress, social support, and the
buffering hypothesis. Psychological Bulletin, 98, 310-357.
Connolly, J. (1976). Life events before myocardial infarction,
Journal of Human Stress, 2, 3-17.
Corey, L., & Holmes, K. K. (1983). Genital herpes simplex virus
infections: Current concepts in diagnosis, therapy, and
prevention. Annals of Internal Medicine, 98, 973-983.
Corey, L., Reeves, W. C., & Holmes, K. K. (1978). Cellular immune
response in genital herpes simplex virus infection. New
England Journal of Medicine, 299, 986-991.
Craig, T. K., & Brown, G. W. (1984). Goal frustrating aspects of
life event stress in the etiology of gastrointestinal
disorder. Journal of Psychosomatic Research, 28, 411-421.
Creed, F. (1981, June). Life events and appendicectomy. Lancet, 1
(8235) 1381-1385.
Creed, F. (1985). Invited review: Life events and physical illness.
Journal of Psychosomatic Research, 29, 113-123.

101
de Araujo, G., Van Arsdel, P. P., Holmes, J. H., & Dudley, D. L.
(1973). Life change, coping ability and chronic intrinsic
asthma. Journal of Psychosomatic Research, 17, 359-363.
DeLongis, A., Coyne, J. C., Dakof, G., Folkman, S., & Lazarus, R. S.
(1982). Relationship of daily hassles, uplifts, and major
life events to health status. Health Psychology, 1, 119-136.
Dohrenwend, B. P. (1974). Problems in defining and sampling the
relevant population of stressful life events. In B.S.
Dohrenwend and B.P. Dohrenwend (Eds.), Stressful life
events: Their nature and effects (pp. 275-310). New York:
John Wiley and Sons.
Dohrenwend, B. S., & Dohrenwend, B. P. (1974). Overview and
prospects for research on stressful life events. In B.S.
Dohrenwend and B.P. Dohrenwend (Eds.), Stressful life
events: Their nature and effects (pp. 313-331). New York:
John Wiley and Sons.
Douglas, J. M., Critchlow, C., Benedetti, J., Mertz, G. J., Connor,
J. D., Hintz, M. A., Fahnlander, A., Pemington, M., Winter,
C., & Corey, L. (1984). A double-blind study of oral
acyclovir for suppression of recurrences of genital herpes
simplex virus infection. New England Journal of Medicine,
310, 1551-1556.
Friedman, S., Katcher, A. H., & Brightman, V. J. (1977). Incidence
of recurrent herpes labial is and upper respiratory infection:
A prospective study of the influence of biologic, social, and
psychologic predictors. Oral Surgery, 43, 873-878.
Funch, D. P., & Marshall, J. R. (1984). Measuring life stress:
Factors affecting fall-off in the reporting of life events.
Journal of Health and Social Behavior, 25, 453-464.
Goldmeier, D., & Johnson, A. (1982). Does psychiatric illness affect
the recurrence rate of genital herpes? British Journal of
Venereal Disease, 58, 40-43.
Gorsuch, R. L., & Key, M. K. (1974). Abnormalities of pregnancy as a
function of anxiety and life stress. Psychosomatic Medicine,
36, 352-361.
Gortmaker, S. L., Eckenroade, J., & Gore, S. (1982). Stress and the
utilization of health services: A time series and cross-
sectional analysis. Journal of Health and Social Behavior,
23, 25-28.
Graves, T. D., & Graves, N. B. (1985). Stress and health among
Polynesian migrants to New Zealand. Journal of Behavioral
Medicine, 8, 1-19.

102
Green, M. T., Rosborough, J. P., & Dunkel, E. C. (1981). In vivo
reactivation of herpes simplex virus in rabbit trigeminal
ganglia: Electrode model. Infection and Immunity,
34(1),69-74.
Hall, N. R., & Goldstein, A. L. (1981). Neurotransmitters and the
immune system. In R. Ader (Ed.), Psychoneuro immunology
(pp. 521-543). New York: Academic Press.
Hamilton, R. D. (1980). The herpes book. Los Angeles: Tarcher,
Inc.
Harney, M. K., & Brigham, T. A. (1984). Tolerance of aversive
stimuli in relation to life change. Journal of Behavioral
Medicine, 8, 21-34.
Harrell, F. E., Jr. (1983). The LOGIST procedure. In S.P. Joyner
(Ed.), SUGI Supplemental library user's guide, 1983 edition
(pp. 63-67). New York: McGraw-Hill.
Hawkins, N. G., Davies, R., & Holmes, T. H. (1957). Evidence of
psychosocial factors in the development of pulmonary
tuberculosis. American Review of Tuberculosis and Pulmonary
Diseases, 75, 768-780.
Help membership HSV survey research project results. (1981). Helper,
3(2), 1-5.
Hill, T.J., & Blyth, W.A. (1976). An alternative theory of herpes
simplex recurrence and a possible role for prostaglandins.
Lancet, 1(7956), 397-398.
Hill, T. J., Blyth, W. A., Harbour, D. A., Berrie, E. L., & Tullo, A.
B. (1983). Latency and other consequences of infection of
the nervous system with herpes simplex virus. In P.0. Behan,
V. ter Meulen, & F.C. Rose (Eds.), Immunology of nervous
system infections, Progress in brain research (Vol. 59, pp.
173-184). New York: Elsevier Science Publishers.
Holmes, T. H., & Masuda, M. (1974). Life change and illness
susceptibility. In B.S. Dohrenwend and B.P. Dohrenwend
(Eds.), Stressful life events: Their nature and effects
(pp. 45-72). New York: John Wiley and Sons.
Holmes, T. H., & Rahe, R. H. (1967). The social readjustment rating
scale. Journal of Psychosomatic Research, 11, 213-218.
Hotaling, G. T., Atwell, S. G., & Linsky, A. S. (1978). Adolescent
life changes and illness: A comparison of three models.
Journal of Youth and Adolescence, 7, 393-403.

103
Hurst, M. W., Jenkins, C. D., & Rose, R. M. (1976). The relation of
psychological stress to onset of medical illness. Annual
Review of Medicine, 27, 301-312.
Jacobs, T. J., & Charles, E. (1980). Life events and the occurrence
of cancer in children. Psychosomatic Medicine, 42, 11-24.
Jacobson, A. M., Rand, L. I., & Hauser, S. T. (1985). Psychologic
stress and glycemic control: A comparison of patients with
and without proliferative diabetic retinopathy.
Psychosomatic Medicine, 47, 372-381.
Janicki, M. P. (1971). Recurrent herpes labialis and recurrent
aphthous ulcerations: Psychological components.
Psychotherapy Psychomatics, 19, 288-294.
Jenkins, C. D., Hurst, M. W., & Rose, R. M. (1979). Life changes:
Do people really remember? Archives of General Psychiatry,
36, 379-384.
Johnson, J. H., & Sarason, I. G. (1978). Life stress, depression,
and anxiety: Internal-external control as a moderator
variable. Journal of Psychosomatic Research, 22, 205-208.
Johnson, J. H., & Sarason, I. G. (1984). Life stress research:
Where have we been—where are we going? Unpublished
manuscript. University of Florida, Department of Clinical
Psychology, Gainesville, FL.
Johnson, J. H., Sarason, I. G., & Siegel, J. M. (1979). Arousal
seeking as a moderator of life stress. Perceptual and Motor
Skills, 49, 665-666.
Kanner, A. D., Coyne, J. C., Schaefer, C., & Lazarus, R. S. (1981).
Comparison of two modes of measurement: Daily hassles and
uplifts versus major life events. Journal of Behavioral
Medicine, 4, 1-39.
Kasl, S. V., Evans, A. S., James, C., & Niederman, J. C. (1979).
Psychosocial risk factors in the development of infectious
mononucleosis. Psychosomatic Medicine, 41, 445-466.
Katcher, A. H., Brightman, V., Luborsky, L., & Ship, I. (1973).
Prediction of the incidence of recurrent herpes labialis and
systemic illness from psychological measurements. Journal of
Dental Research, 52, 49-58.
Keddie, F. M., Reeves, R. B., & Epstein, N. N. (1941). Herpes
simplex following artificial fever therapy. Journal of the
American Medical Association, 117, 1327-1330.

104
Kern, A. (1979). Recurrent herpes simplex: A simple treatment.
Consultant, 19, 83.
Kiecolt-Glaser, J. K., Ricker, D., George, J., Messick, G., Speicher,
C. E., Garnea, W., & Glaser, R. (1984). Urinary cortisol
levels, cellular immunocompetency and loneliness.
Psychosomatic Medicine, 46, 15-23.
Kirchner, H., Schwenteck, M., Northoff, H., & Schopf, E. (1978).
Defective in vitro lymphoproliterative responses to herpes
simplex virus in patients with frequently recurring herpes
infections during the disease-free interval. Clinical
Immunology and Immunopatholog.y, 11, 267-274.
Kobasa, S. C. (1979). Stressful life events, personality, and
health: An inquiry into hardiness. Journal of Personality
and Social Psychology, 37, 1-11.
Kobasa, S. C., Maddi, S. R., & Courington, S. (1981). Personality
and constitution as mediators in the stress-illness
relationship. Journal of Health and Social Behavior, 22,
368-378.
Laibson, P. R., & Kibrick, S. (1966). Reactivation of herpetic
keratitis by epinephrine in rabbit. Archives of Opthamology,
75, 254-260.
Lai, N., Ahuja, R. C., & Madhukar, A. (1982). Life events in
hypertensive patients. Journal of Psychosomatic Research,
26, 441-445.
Levenson, H., Hirschfeld, M. L., Hirschfeld, A., & Dzubay, B. (1983).
Journal of Human Stress, 9, 4-8.
Liao, W. (1977). Psychological stress, health status, and health
behavior: Application of the life changes concept.
Psychological Reports, 41, 246.
Lin, N., Ensel, W. M., Simeone, R. S., & Kuo, W. (1979). Social
support, stressful life events and illness: A model and an
empirical test. Journal of Health and Social Behavior, 20,
108-119.
Linn, M. W. (1982). In defense of statistical significance with
multivariate analysis. Diabetes Care, 5, 553-554.
Linn, M. W., Linn, B. S., Skyler, J. S., & Jensen, J. (1983). Stress
and immune function in diabetes mellitus. Clinical
immunology and immunopathology, 27, 223-233.
Locke, S. E. (1982). Stress, adaptation and immunity; Studies in
humans. General Hospital Psychiatry, 4, 49-58.

105
Luborsky, L., Mintz, J., Brightman, V. J., & Katcher, A. H. (1976).
Herpes simplex and moods: A longitudinal study. Journal of
Psychosomatic Research. 20, 543-548.
Luby, E. D., & Klinge, V. (1985). Genital herpes: A pervasive
psychosocial disorder. Archives of Dermatoloqy, 121,
494-497.
Manuck, S. B., Hinrichsen, J. J., & Ross, E. 0. (1975). Life stress,
locus of control and treatment seeking. Psychological
Reports, 37, 589-590.
Marx, M. B., Garrity, R. F., & Bowers, F. R. (1975). The influence
of recent life experience on the health of college freshmen.
Journal of Psychosomatic Research, 19, 87-98.
McDowell, M. (1959). Hypnosis in dermatology. In J.M. Schneck
(Ed.), Hypnosis in modern medicine (2nd ed., pp. 101-115).
Springfield, IL: Charles C. Thomas.
McFarlane, A. H., Norman, G. R., Streiner, D. L., & Roy, R. G.
(1983). The process of social stress: Stable, reciprocal,
and mediating relationships. Journal of Health and Social
Behavior, 24, 160-173.
Mechanic, D. (1975). Some problems in the measurement of stress and
social readjustment. Journal of Human Stress, 1, 43-48.
Mehrabian, A. (1978). Characteristic individual reactions to
preferred and unpreferred environments. Journal of
Personal it.y, 46, 717-731.
Mehrabian, A., & Russell, J. A. (1974). An approach to environmental
ps.ychology (pp. 29-55). Cambridge, MA: MIT Press.
Mertz, G., & Corey, L. (1984). Genital herpes simplex virus
infections in adults. Urologic Clinics of North America,
11(1), 103-119.
Meyer, R. J., & Haggerty, R. J. (1962). Streptococcal infection in
families: Factors in the onset of somatic illness in
children. Pediatrics, 29, 539-549.
Meyers, J. D., Flournoy, N., & Thomas, E. D. (1980). Infection with
herpes simplex virus and cell-mediated immunity after marrow
transplant. Journal of Infectious Diseases, 142, 338-346.
Miller, P.M., Ingham, J.G., & Davidson, S. (1976). Life events,
symptoms, and social support. Journal of Psychosomatic
Research, 20, 515-522.
Minter, R. E., & Kimball, C. P. (1978). Life events and illness
onset: A review. Psychosomatics, 19, 334-339.

106
Monroe, S. M. (1982a). Assessment of life events: Retrospective vs.
concurrent strategies. Archives of General Psychiatry, 39,
606-610.
Monroe, S. M. (1982b). Life events assessment: Current practices,
emerging trends. Clinical Psychology Review, 2, 435-453.
Mueller, D. P., Edwards, D. W., & Yarvis, R. M. (1977). Stressful
life events and psychiatric symptomatology: Change as
undesirability. Journal of Health and Social Behavior, 18,
307-317.
Murphy, E., & Brown, G. W. (1980). Life events, psychiatric
disturbance and physical illness. British Journal of
Psychiatry, 136, 326-338.
Nahmias, A. J., & Roizman, B. (1973a). Infection with herpes-simplex
viruses 1 and 2 (Part 1). New England Journal of Medicine,
289(13), 667-674.
Nahmias, A. J., & Roizman, B. (1973b). Infection with herpes-simplex
viruses 1 and 2 (Part 2). New England Journal of Medicine,
289(14), 719-725.
Nahmias, A. J., & Roizman, B. (1973c). Infection with herpes-simplex
viruses 1 and 2 (Part 3). New England Journal of Medicine,
289(15), 781-789.
Newton, R. W., & Hunt, L. P. (1984). Psychosocial stress in
pregnancy and its relation to low birth weight. British
Medical Journal, 288, 1191-1194.
Norbeck, J. S., & Tilden, V. P. (1983). Life stress, social support
and emotional disequilibrium in complications of pregnancy:
A prospective, multivariate study. Journal of Health and
Social Behavior, 24, 30-45.
Nuckolls, K. B., Cassel, J., & Kaplan, B. H. (1972). Psychosocial
assets, life crisis and the prognosis of pregnancy. American
Journal of Epidemiology, 95, 431-441.
Parkes, M. C., Benjamin, B., & Fitzgerald, R. G. (1969). Broken
heart: A statistical study of increased mortality among
widowers. British Medical Journal, 1, 740-743.
Pass, R. F., Whitley, R. J., Whelchel, J. D., Diethelm, A. G.,
Reynolds, D. W., & Alford, C. A. (1979). Identification of
patients with increased risk of infection with herpes simplex
virus after renal transplantation. The Journal of Infectious
Diseases, 140, 487-492.

107
Passer, M. W., & Seese, M. D. (1983, December). Life stress and
athletic injury: Examination of positive versus negative
events and three moderator variables. Journal of Human
Stress, 9(4), 11-16.
Paykel, E. S. (1983). Methodological aspects of life events
research. Journal of Psychosomatic Research, 27, 341-352.
Pazin, G. J., Armstrong, J. A., Lam, M. T., Tarr, G. C., Janetta, P.
J., & Ho, M. (1979). Prevention of reactivated herpes
simplex infection by human leukocyte interferon after
operation on the trigeminal root. New England Journal of
Medicine, 301, 225-230.
Plaut, S. M., & Friedman, S. B. (1981). Psychosocial factors in
infectious disease. In R.E. Ader (Ed.),
Psychoneuroimmunology (pp.3-30). New York: Academic Press.
Rabkin, J. G., & Struening, E. L. (1976). Life events, stress and
illness. Science, 194, 1013-1020.
Rahe, R. H., & Lind, E. (1971). Psychosocial factors and sudden
cardiac death: A pilot study. Journal of Psychosomatic
Research, 15, 19-24.
Rahe, R. H., Mahan, J. L., & Arthur, R. J. (1970). Prediction of
near future health changes from subjects preceding life
changes. Journal of Psychosomatic Research, 14, 401-406.
Rand, K. H. (1982). Herpes simplex virus: Clinical syndromes and
current therapy. Comprehensive Therapy, 8(2), 44-50.
Rand, K. H., Kramer, B., & Johnson, A. C. (1982). Cancer
chemotherapy associated symptomatic stomatitis: Role of
herpes simplex virus (HSV). Cancer, 50(7), 1262-1265.
Rand, K. H., Rasmussen, L. E., Pollard, R. B., Arvin, A., & Merigan,
T. C. (1977). Cellular immunity and herpes virus infections
in cardiac transplant patients. New England Journal of
Medicine, 296, 1372-1377.
Rizzardo, R., Magni, G., Andreoli, C., Merlin, G., Andreoli, F.,
Consentino, M., & Ziglio, R. (1982). Life events and
obstetrical complications: A preliminary report.
Psychological Reports, 51, 1035-1038.
Rogers, M. P., Dubey, D., & Reich, P. (1979). The influences of the
psyche and the brain on immunity and disease susceptibility:
A critical review. Psychosomatic Medicine, 41, 147-164.
Rola-Plezczynski, M., & Lieu, H. (1984). Natural cytotoxic cell
activity linked to time of recurrence of herpes labialis.
Clinical and Experimental Immunology, 55, 224-228.

108
Roth, D. L., & Holmes, D. S. (1985). Influence of physical fitness
in determining the impact of stressful life events on
physical and psychologic health. Psychosomatic Medicine, 47,
164-173.
Rotter, J. B. (1966). Generalized expectancies for internal vs.
external control of reinforcement. Psychological Monographs,
80 (1, Serial No. 609).
Sarason, I. G., Johnson, J. H., & Siegel, J. M. (1978). Development
of the life experiences survey. Journal of Consulting and
Clinical Psychology, 46, 932-946.
Sarason, I. G., Levine, H. M., Basham, R. B., & Sarason, B. R.
(1981). Assessing social support: The social support
questionnaire (NR Report No. C0-004). Arlington, VA: Office
of Naval Research.
Sarason, I. G., Levine, H. M., & Sarason, B. R. (1982). Assessing
the impact of life changes. In T. Millón, C. Green, and R.
Meagher (Eds.), Handbook of clinical health psychology (pp.
377-399). New York: Plenum.
SAS user's guide Basics (5th ed. 1985). Cary, NC: SAS Institute.
Schaefer, C., Coyne, J. C., & Lazarus, R. S. (1981). The
health-related functions of social support. Journal of
Behavioral Medicine, 4, 381-406.
Schneck, J. M. (1947). The psychological component in a case of
herpes simplex. Psychosomatic medicine, 9, 62-64.
Schwartz, G. E. (1982). Testing the biopsychosocial model: The
ultimate challenge facing behavioral medicine? Journal of
Consulting and Clinical Psychology, 50, 1040-1053.
Sheehan, D. V., O'Donnell, J., Fitzgerald, A., Harvey, L., & Ward, H.
(1979). The relationship between response qualities of life
change events and future illness rates. International
Journal of Psychiatry in Medicine, 9, 217-225.
Siegel, J. M., Johnson, J. H., & Sarason, I. G. (1979). Life stress
and menstrual discomfort. Journal of Human Stress, 5, 41-46.
Siegel, S. (1956). Nonparametric statistics for the behavioral
sciences (pp. 63-67). New York: McGraw-Hill.
Skinner, H. A., & Lei, H. (1980). Differential weights in life
change research: Useful or irrelevant? Psychosomatic
Medicine, 42, 367-370.

109
Smith, R. E., Johnson, J. H., & Sarason, I. G. (1978). Life change,
the sensation seeking motive, and psychological distress.
Journal of Consulting and Clinical Psychology, 46, 348-349.
Stevens, J. G. (1975). Latent herpes simplex virus and the nervous
system. Current Topics in Microbiology and Immunology, 70,
31-50.
Straus, S. E., Takiff, H. E., Seidlin, M., Bachrach, S., Linninger,
L., DiGiovanna, J. J., Western, K. A., Smith, H. A., Lehrman,
S. N., Creagh-Kirk, T., & Ailing, D. W. (1984). Suppression
of frequently recurring genital herpes: A placebo-controlled
double-blind trial of oral acyclovir. New England Journal of
Medicine, 310, 1545-1550.
Taylor, B. J. (1978). The psychological and behavioral effects of
genital herpes in women: High recurrers vs. low recurrers
(Doctoral dissertation, University of Washington 1978).
Dissertation Abstracts International, 39(5-B), 2529-2530.
(University Microfilms No. 7820779)
Tessler, R., Mechanic, D., & Dimond, M. (1976). The effect of
psychological distress on physician utilization: A
prospective study. Journal of Health and Social Behavior,
17, 353-364.
Theorell, T., & Rahe, R. H. (1971). Psychosocial factors and
myocardial infarction. I: An inpatient study in Sweden.
Journal of Psychosomatic Research, 15, 25-31.
Thoits, P. A. (1982). Conceptual, methodological and theoretical
problems in studying social support as a buffer against life
stress. Journal of Health and Social Behavior, 23, 145-159.
Thurlow, H. J. (1971). Illness in relation to life situation and
sick role tendency. Journal of Psychosomatic Research, 15,
73-88.
Totman, R., Kiff, J., Reed, S. E., & Craig, J. W. (1980). Predicting
experimental colds in volunteers from different measures of
recent life stress. Journal of Psychosomatic Research, 24,
155-163.
Toves, C., Schill, T., & Ramanaiah, N. (1981). Sex differences,
internal-external control and vulnerability to life stress.
Psychological Reports, 49, 508.
Ullman, M. (1947). Herpes simplex and second degree burn induced
under hypnosis. American Journal of Psychiatry, 103,
828-830.

no
Vinokur, A., & Selzer, M. L. (1975). Desirable versus undesirable
life events: Their relationship to stress and mental
distress. Journal of Personality and Social Psychology, 32,
329-337.
Watson, D. B. (1983). The relationship of genital herpes and life
stress as moderated by locus of control and social support
(Doctoral dissertation, University of Southern California,
1983). Dissertation Abstracts International, 43(A-12),
3857-3858.
Wickett, W. H. (1982). Herpes: Cause and control. New York:
Pinnacle Books.
Williams, N. A., & Deffenbacher, J. L. (1983, December). Life stress
and chronic yeast infections. Journal of Human Stress, 9(1),
26-31.
Wyler, A. R., Masuda, M., & Holmes, J. H. (1971). Magnitude of life
events and seriousness of illness. Psychosomatic Medicine,
33, 115-122.
Yaeger, A. S., Morris, J. E., & Prober, C. G. (1977). Storage and
transport of cultures for herpes simplex virus type 2.
American Journal of Clinical Pathology, 72, 977.
Zimmerman, M. (1983). Weighted versus unweighted life event scores:
Is there a difference? Journal of Human Stress, 9(4), 30-35.
Zuckerman, M. (1971). Dimensions of sensation seeking. Journal of
Consulting and Clinical Psychology, 36, 45-52.

BIOGRAPHICAL SKETCH
Emily Franck Hoon was born and grew up with her parents and only
sibling, a younger sister, in Connecticut. After completing high
school in Connecticut, she attended Grinnell College in Iowa where she
received a B.A. degree in psychology. Emily involved herself for 2
years teaching educable mentally retarded children in New York and
Connecticut. She continued for another 2 years teaching as a Peace
Corps volunteer stationed in Mindinao, the Philippines, with her
husband, Pete. The following 4 years were spent in Lincoln, Nebraska,
where Emily served as a director of a women's residence hall at the
University of Nebraska while Pete completed a degree in counseling
psychology.
Wendy, the first of two children, was born in Lincoln. She was
joined by her brother, Robin, a year and a half later while the family
was in Kansas City, Kansas. The family spent some time in Ponce,
Puerto Rico, while Pete consulted for a Peace Corps training
project. They then resided for a year in southwestern Pennsylvania
followed by another 1 year stay in Stony Brook, New York. The family
finally settled down for a 4-year period in Halifax, Nova Scotia,
Canada. The children were old enough for Emily to branch out.
Emily developed her reputation as a sex therapist while
functioning as a co-therapist and researcher at the Clinical Research
Center at Dalhousie University. A string of publications to her
111

112
credit have grown from this beginning. While serving as a part-time
member of the Psychology Department at the Nova Scotia Provincial
Hospital, Emily earned her master's degree in clinical psychology at
Acadia University. This was followed by 3 years of valuable
experience in the Psychiatry Department of the University of Tennessee
Center for Health Sciences where Emily served as a researcher,
clinician, and instructor.
Desiring further education, Emily committed her energies toward
the acquisition of her Ph.D. in clinical psychology at the University
of Florida with a special interest in medical psychology. One might
get the impression that Emily has spent all of her time acquiring
education and performing in her areas of specialization. However, her
family and friends see her as a life-loving, energetic, enthusiastic
participant in scuba diving, swimming, river running (loves white
water), camping, hiking, and dancing.
Emily enjoys and appreciates people. In addition to enjoying the
interaction with people in the pursuit of her education and the
practice of her profession, Emily shares a good portion of her time in
a variety of activities with the people of the Unitarian Universalist
Fellowship of Gainesville, Florida, of which she is a member and with
friends she has made in other nonprofessional memberships.
Should Emily continue in her current trend, we can expect to hear
more from her.

I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.
\ C\
>-Q- U-
James H. Johnson Chairman
Associate Professor of Clinical
Psychology
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.
Randy Carte/
Associate Professor of Statistics
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.
¿lUl^ Qh 3(j^kjJJL
Eileen B. Fennell
Associate Professor of Clinical
Psychology
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.
flfiKAA TUTuaj? I g
Nancy Norvdjl
Assistant Professor of Clinical
Psychology

I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is
fully adequate, in scope and quality, as a dissertation for the degree
of Doctor of Philosophy.
Associate Professor of Clinical
Psychology
This dissertation was submitted to the Graduate Faculty of the College
of Health Related Professions and to the Graduate School and was
accepted as partial fulfillment of the requirement for the degree of
Doctor of Philosophy.
May, 1986
. f.l
iñpan Pnliphp nr Hpa1
¿uJLJ ,r
'Dean, College of Health Related
Professions
Dean, Graduate School

UNIVERSITY OF FLORIDA
3 1262 08554 3964




PAGE 1

/,)( 675(66 ,03$&7 21 *(1,7$/ +(53(6 5(&855(1&(6 %< (0,/< )5$1&. +221 $ ',66(57$7,21 35(6(17(' 72 7+( *5$'8$7( 6&+22/ 2) 7+( 81,9(56,7< 2) )/25,'$ ,1 3$57,$/ )8/),//0(17 2) 7+( 5(48,5(0(176 )25 7+( '(*5(( 2) '2&725 2) 3+,/2623+< 81,9(56,7< 2) )/25,'$

PAGE 2

$&.12:/('*0(176 7KLV SURMHFW LQYROYHG WKH WLPH WDOHQW HIIRUW DQG FRRSHUDWLRQ RI PDQ\ LQGLYLGXDOV %HFDXVH DSSUHFLDWH WKHLU FRQWULEXWLRQV DQG ZDQW WR UHFRJQL]H WKHP DQG WKDQN WKHP DGGUHVV WKH IROORZLQJ .HQ 5DQGf§SDWLHQW HQWKXVLDVWLF KXPRURXV HQFRXUDJLQJ NQRZOHGJDEOH DFFHVVLEOHf§IULHQG FRQWULEXWHG IUHHO\f§UHPHPEHUHG ZLWK ZDUPWK IRUHYHU -LP -RKQVRQf§JXLGDQFH VHQVH RI RUJDQL]DWLRQ SHUVSHFWLYH FDOPLQJ SUHVHQFH ZLVGRPf§D UROH PRGHO GHVHUYLQJ HPXODWLRQf§P\ VLQFHUH WKDQNV 5DQG\ &DUWHU DQG KLV DVVLVWDQW VWDWLVWLFLDQV ZKR KHOSHG PDNH VHQVH RI YROXPLQRXV FRPSXWHU SULQWRXWV $QG WR WKH UHVW RI WKH EHVW FRPPLWWHH D 3K' FDQGLGDWH HYHU KDG (LOHHQ )HQQHOO 5XG\ 9XFKLQLFK DQG 1DQF\ 1RUYHOO :HQG\ DQG 5RELQ +RRQf§\RXQJ SHRSOH JURZQ LQGHSHQGHQW EHIRUH P\ H\HV RXW RI QHFHVVLW\ ZKLOH FRSLQJ ZLWK WKH YDJDULHV RI WKHLU 0RPnV JUDGXDWH VFKRRO VFKHGXOHf§, ORYH ERWK RI \RX DQG WKDQN \RX IRU \RXU KHOS DQG FRRSHUDWLRQ (UQHVW )UDQFN P\ GDG ZKR KDV DOZD\V HQFRXUDJHG PH WR IROORZ P\ FXULRVLW\ DQG WR DFFHSW FKDOOHQJHV 7KH ODWH (PLO\ )UDQFN P\ UQRP ZKR KHOSHG PH WR DSSUHFLDWH WKH YDOXH RI VRFLDO UHODWLRQVKLSV

PAGE 3

3HWH +RRQ P\ H[KXVEDQG ZLWK ZKRP VKDUHG P\ HDUO\ UHVHDUFK H[SHULHQFHV DQG ZLWK ZKRP FRQWLQXH WR HQMR\ D VDWLVI\LQJ FROODERUDWLYH UHODWLRQVKLS DV D SURIHVVLRQDO DQG FRSDUHQW 5RE 0DUWLQ P\ SUHVHQW KXVEDQG ZKR KHOSHG PH LQ XQFRXQWDEOH ZD\V WR WLH XS ORRVH HQGV DQG ZKR SURYLGHV LQVSLUDWLRQ IRU JHWWLQJ WKH PRVW RXW RI ZRUN WLPH DQG SOD\ WLPH 6XEMHFWVf§FRXUDJHRXVO\ VKDULQJ SHUVRQDO LQIRUPDWLRQ WR DGYDQFH NQRZOHGJH UHJDUGLQJ KHUSHV $QG WKH IROORZLQJ FKHHUIXO KHOSHUV UHVHDUFK DVVLVWDQWV ODERUDWRU\ SHUVRQQHO DQG &OLQLFDO 5HVHDUFK &HQWHU VWDII L L L

PAGE 4

7$%/( 2) &217(176 3DJH $&.12:/('*0(176 LL $%675$&7 YL &+$37(56 21( /,7(5$785( 5(9,(: 6WUHVV DQG ,OOQHVV 0RGHUDWRU 9DULDEOHV 6RFLDO 6XSSRUW /RFXV RI &RQWURO $URXVDO 6HHNLQJ 0HWKRGRORJLFDO ,VVXHV 5HWURVSHFWLYH 'HVLJQ 0HDVXUHV RI ,OOQHVV &RQIRXQGLQJ RI (YHQWV DQG 2XWFRPH 0HDVXUHV 9DOXH DQG ,PSDFW RI (YHQWV 0LQRU (YHQWV 5HFXUUHQW *HQLWDO +HUSHV DV D 0RGHO &OLQLFDO 1DWXUH RI +HUSHV 6LPSOH[ 9LUXV 5HFXUUHQFHV 3UHVHQW /DFN RI &XUH /DERUDWRU\ 'RFXPHQWDWLRQ (QGRJHQRXV 3DWKRJHQ 6WUHVV DQG +HUSHV $QHFGRWDO 5HSRUWV 6XUYH\ 5HVHDUFK 5HWURVSHFWLYH 5HVHDUFK 3URVSHFWLYH 5HVHDUFK 6XPPDU\ 7KHRUHWLFDO )UDPHZRUN 3HUFHSWLRQ RI 6WUHVVIXO (YHQWV 3URSRVHG 3K\VLRORJLFDO 0HFKDQLVPV 3UHVHQW ,QYHVWLJDWLRQ +\SRWKHVHV 7:2 0(7+2'6 5HVHDUFK 'HVLJQ 6DPSOH 6HOHFWLRQ L Y

PAGE 5

0HDVXUHV 6WUHVV 0HDVXUHV /LIH ([SHULHQFHV 6XUYH\ /(6f +DVVOHV 6FDOH +6f 0RGHUDWRU 9DULDEOHV /RFXV RI &RQWURO /&f $URXVDO 6HHNLQJ $6f 6RFLDO 6XSSRUW 4XHVWLRQQDLUH 664f 3K\VLRORJLFDO 0HDVXUHV 6FUHHQLQJ SURFHGXUH +69 DQWLERG\ 'HSHQGHQW YDULDEOH 9LUXV LVRODWLRQ DQG WUDQVSRUW 3URFHGXUH ,QLWLDO &RQWDFW 6HFRQG &RQWDFW &RQWLQXLQJ &RQWDFW )LQDO &RQWDFW 7+5(( 5(68/76 6DPSOH &KDUDFWHULVWLFV 8QGHVLUDEOH /LIH (YHQWV 3UHVXPHG 0RGHUDWRU 9DULDEOHV 'HILQLWLRQ RI 5HFXUUHQFH 6WUHVV5HFXUUHQFH 5HODWLRQVKLS ,QIOXHQFH RI 0RGHUDWRU 9DULDEOHV 7KH 1DWXUH RI WKH /LIH 6WUHVV5HFXUUHQFH 5HODWLRQVKLS )285 ',6&866,21 $33(1',&(6 $ 6758&785(' ,17(59,(: 68%-(&76n 727$/ 6&25(6 21 675(66 02'(5$725 $1' 5(&855(1&( 9$5,$%/(6 & 63($50$1 &255(/$7,21 &2()),&,(176 $1' 6,*1,),&$1&( /(9(/6 )25 5(3257(' 675(66 )5(48(1&< $1' 5(&855(1&( '(),1,7,216 5()(5(1&(6 %,2*5$3+,&$/ 6.(7&+ ,OO Y

PAGE 6

$EVWUDFW RI 'LVVHUWDWLRQ 3UHVHQWHG WR WKH *UDGXDWH 6FKRRO RI WKH 8QLYHUVLW\ RI )ORULGD LQ 3DUWLDO )XOILOOPHQW RI WKH 5HTXLUHPHQWV IRU WKH 'HJUHH RI 'RFWRU RI 3KLORVRSK\ /,)( 675(66 ,03$&7 21 *(1,7$/ +(53(6 5(&855(1&(6 %< (0,/< )5$1&. +221 0D\ &KDLUPDQ -DPHV + -RKQVRQ 0DMRU 'HSDUWPHQW &OLQLFDO 3V\FKRORJ\ 7KH SUHVHQW VWXG\ RI WKH HIIHFWV RI PDMRU DQG PLQRU OLIH HYHQWV RQ UHFXUUHQFHV RI JHQLWDO KHUSHV DGGUHVVHG WKH QHHG IRU SURVSHFWLYH UHVHDUFK ZLWK REMHFWLYH LQGLFHV RI LOOQHVV LH YLUDO FXOWXUHf DQG SURYLGHG HPSLULFDO LQIRUPDWLRQ UHJDUGLQJ D VWUHVVUHFXUUHQFH OLQN )RU PRQWKV +69 VHURSRVLWLYH VXEMHFWV PRQLWRUHG VWUHVV DQG KDG DOO UHFXUUHQW OHVLRQV FXOWXUHG 6XEMHFWV ZHUH DGPLQLVWHUHG WKH /LIH ([SHULHQFHV 6XUYH\ /(6f XSRQ HQUROOPHQW UHJDUGLQJ PDMRU HYHQWV H[SHULHQFHG RYHU WKH SDVW PRQWKV 7KLV LQIRUPDWLRQ ZDV DQDO\]HG WR GHWHUPLQH WKH H[WHQW RI D SURVSHFWLYH UHODWLRQVKLS WR VXEVHTXHQW UHFXUUHQFHV 7KHUHDIWHU WKH /(6 ZDV DGPLQLVWHUHG PRQWKO\ ZLWK WKH +DVVOHV 6FDOH +6f ZKLFK FRQFHUQV PLQRU VWUHVVRUV 7R LGHQWLI\ SHUVRQDO FKDUDFWHULVWLFV ZKLFK PLJKW EH UHODWHG WR IUHTXHQW UHFXUUHQFHV XQGHU FRQGLWLRQV RI VWUHVV WKH IROORZLQJ SUHVXPHG PRGHUDWRU YDULDEOHV ZHUH H[DPLQHG VRFLDO VXSSRUW ORFXV RI FRQWURO RULHQWDWLRQ DQG DURXVDO VHHNLQJ WHQGHQF\ YL

PAGE 7

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f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

PAGE 8

&+$37(5 21( /,7(5$785( 5(9,(: 6WUHVV DQG ,OOQHVV 7KH QRWLRQ RI D OLQN EHWZHHQ SV\FKRVRFLDO IDFWRUV DQG LOOQHVV LV QRW QHZ 3K\VLFLDQV KDYH SUHVFULEHG HQYLURQPHQWDO DQG EHKDYLRUDO LQWHUYHQWLRQV IRU \HDUV EDVHG XSRQ WKHLU FOLQLFDO H[SHULHQFH DQG LQWXLWLRQ :KDW LV QHZ LV D IOXUU\ RI DFWLYLW\ GHVLJQHG WR GRFXPHQW DQG FODULI\ WKLV UHODWLRQVKLS :LWK UHJDUG WR OLIH VWUHVV DV D SV\FKRVRFLDO IDFWRU LQ WKH HWLRORJ\ RI LOOQHVV +ROPHV DQG 5DKH f FDQ EH FUHGLWHG ZLWK LQLWLDWLQJ D QHZ ZDYH RI UHVHDUFK ZLWK WKH GHYHORSPHQW RI WKHLU PHDVXUH WKH 6FKHGXOH RI 5HFHQW ([SHULHQFH 65(f 7KLV ZDV WKH ILUVW DWWHPSW WR TXDQWLI\ XVLQJ D VLPSOH LWHP TXHVWLRQQDLUH IRUPDW WKH VWUHVVIXO OLIH HYHQWV EHOLHYHG WR KHLJKWHQ YXOQHUDELOLW\ WR LOOQHVV 7KH 65( DQG YDULRXV PRGLILFDWLRQV RI WKLV PHDVXUH KDYH EHHQ ZLGHO\ XVHG WR SURYLGH HPSLULFDO HYLGHQFH IRU WKH UHODWLRQVKLS EHWZHHQ OLIH VWUHVV GHILQHG LS WHUPV RI PDMRU OLIH FKDQJHVf DQG LOOQHVV &RQVLVWHQW EXW PRGHVW FRUUHODWLRQV KDYH EHHQ FRQILUPHG ZLWK D YDULHW\ RI LOOQHVVHV DQG LOOQHVVUHODWHG PHDVXUHV LQ WKH SDVW WZR GHFDGHV (OHYDWHG OLIH VWUHVV KDV EHHQ VKRZQ WR EH DVVRFLDWHG ZLWK KLJKHU UHSRUWHG LOOQHVV UDWHV DPRQJ PDQ\ JURXSV +ROPHV t 0DVXGD 0DU[ *DUULW\ t %RZHUV 5DKH 0DKDQ t $UWKXU f WKH VHULRXVQHVV RI LOOQHVV :\OHU 0DVXGD t +ROPHV f DWKOHWLF LQMXULHV

PAGE 9

%UDPZHOO 0DVLLGD :DJQHU t +ROPHV f ZRUNUHODWHG DFFLGHQWV /HYHQVRQ +LUVFKIHOG +LUVFKIHOG t ']XED\ f DQG WKH FRPSOLFDWLRQV RI SUHJQDQF\ DQG FKLOGELUWK *RUVXFK t .H\ 1HZWRQ t +XQW 1XFNROOV &DVVHO t .DSODQ 5L]]DUGR HW DO f ,OOQHVVUHODWHG EHKDYLRUV VXFK DV WKH VHHNLQJ RI PHGLFDO DWWHQWLRQ DQG WDNLQJ GLVDELOLW\ GD\V KDYH DOVR EHHQ DVVRFLDWHG ZLWK KLJKHU VFRUHV RQ OLIH FKDQJH LQGLFHV *RUWPDNHU (FNHQURDGH t *RUH /LDR 0LOOHU ,QJKDP t 'DYLGVRQ 0XUSK\ t %URZQ 7HVVOHU 0HFKDQLF t 'LPRQG 7KXUORZ f 6LPLODUO\ WKH RQVHW RI SDUWLFXODU GLVHDVHV KDV EHHQ DVVRFLDWHG ZLWK HOHYDWHG OLIH FKDQJH VWUHVV VFRUHV 7KHVH LQFOXGH P\RFDUGLDO LQIDUFWLRQ &RQQROO\ 7KHRUHOO t 5DKH f VXGGHQ FDUGLDF GHDWK 5DKH t /LQG f FKLOGKRRG FDQFHU -DFREV t &KDUOHV f JDVWURLQWHVWLQDO GLVRUGHUV &UDLJ t %URZQ f DSSHQGLFLWLV &UHHG f WXEHUFXORVLV +DZNLQV 'DYLHV t +ROPHV f PRQRQXFOHRVLV .DVO (YDQV t 1LHGHUPDQ f K\SHUWHQVLRQ /DL $KXMD t 0DGKXNDU f PHQVWUXDO DEQRUPDOLWLHV +DUULV FLWHG LQ &UHHG 6LHJHO -RKQVRQ t 6DUDVRQ f FKURQLF \HDVW LQIHFWLRQV :LOOLDPV t 'HIIHQEDFKHU f VWUHSWRFRFFDO LQIHFWLRQV 0H\HU t +DJJHUW\ f DQG H[SHULPHQWDOO\ LQGXFHG FROG LQIHFWLRQV 7RWPDQ .LII 5HHG t &UDLJ f :KDW LV VWULNLQJ LQ WKHVH H[DPSOHV LV WKH ZLGH UDQJH RI KHDOWKUHODWHG SUREOHPV ZKLFK EHDU D VLJQLILFDQW UHODWLRQVKLS WR PHDVXUHV RI OLIH VWUHVV ,W LV DV WKRXJK H[FHVV VWUHVV LQFUHDVHV RQHnV YXOQHUDELOLW\ WR LOOQHVV LQ JHQHUDO QRW MXVW D IHZ VSHFLILF GLVRUGHUV ,QGHHG WKLV LV FRQVLVWHQW ZLWK WKH YLHZV RI +ROPHV DQG 0DVXGD f WKDW OLIH VWUHVV UHVXOWV LQ LOOQHVV

PAGE 10

WKURXJK LQFUHDVHG VXVFHSWLELOLW\ WR D UDQJH RI KHDOWKUHODWHG SUREOHPV ,Q DGGLWLRQ WR VWXG\LQJ WKH UHODWLRQVKLS EHWZHHQ OLIH VWUHVV DQG LOOQHVV RQVHW RWKHU VWXGLHV KDYH IRFXVHG RQ WKH UHODWLRQVKLS EHWZHHQ OLIH VWUHVV DQG IOXFWXDWLRQV RI KHDOWK VWDWXV LQ LQGLYLGXDOV ZLWK FKURQLF LOOQHVV &KLOGUHQ ZLWK D PL[HG JURXS RI FKURQLF FRQGLWLRQV HJ GLDEHWHV DVWKPD EOLQGQHVVf ZKR DWWHQGHG D ZHHN VXPPHU FDPS ZHUH REVHUYHG E\ %HGHOO *LRUGDQL $UPRXU 7DYRUPLQD DQG %ROO f $ UHODWLRQVKLS EHWZHHQ VHOIUHSRUWHG OLIH FKDQJH DQG IUHTXHQF\ RI LOOQHVVUHODWHG V\PSWRPDWLF HSLVRGHV ZDV IRXQG 6LPLODUO\ DVWKPDWLF LQGLYLGXDOV PRQLWRUHG SURVSHFWLYHO\ ZHUH IRXQG WR UHTXLUH PRUH PHGLFDWLRQ XQGHU FRQGLWLRQV RI LQFUHDVHG OLIH VWUHVV GH $UDXMR 9DQ $UVGHO +ROPHV t 'XGOH\ f 'LDEHWLF FRQWURO KDV DOVR EHHQ LQYHVWLJDWHG %UDGOH\ f IRXQG OLIH VWUHVV WR EH UHODWHG WR LQFUHDVHV LQ EORRG VXJDU WKH QHHG IRU FOLQLF YLVLWV DQG QHFHVVDU\ SUHVFULSWLRQ FKDQJHV SDUWLFXODUO\ IRU LQVXOLQGHSHQGHQW GLDEHWLFV 6LPLODUO\ LQVXOLQGHSHQGHQW GLDEHWLFV ZHUH IRXQG WR EH PRUH YXOQHUDEOH WR OLIH VWUHVV WKDQ DJH DQG VH[PDWFKHG QRQLQVXOLQ GHSHQGHQW GLDEHWLFV DQG PDWFKHG QRQGLDEHWLF FRQWUROV /LQQ /LQQ 6N\OHU -HQVHQ f ,QVXOLQGHSHQGHQW GLDEHWLFV SHUFHLYHG PRUH VWUHVV DQG VKRZHG SRRUHU PHWDEROLF FRQWURO GHVSLWH VLPLODU FRPSOLDQFH ZLWK WKH PHGLFDO UHJLPHQ 7KH\ DOVR GHPRQVWUDWHG GHFUHDVHG LPPXQH UHVSRQVLYLW\ /LQQ HW DO f -DFREVRQ 5DQG DQG +DXVHU f VWXGLHG WKH LPSDFW RI OLIH HYHQWV RQ WKH ORQJ WHUP FRPSOLFDWLRQV RI D JURXS RI LQVXOLQGHSHQGHQW GLDEHWLFV $ VPDOO EXW VLJQLILFDQW UHODWLRQVKLS ZDV IRXQG EHWZHHQ QHJDWLYH OLIH HYHQWV DQG JO\FHPLF

PAGE 11

FRQWURO IRU DOO VXEMHFWV ,Q WKH VXEJURXS RI GLDEHWLFV ZLWK SUROLIHUDWLYH UHWLQRSDWK\ RI UHFHQW RQVHW D VWURQJHU UHODWLRQVKLS ZDV IRXQG EHWZHHQ ODFN RI JO\FHPLF FRQWURO DQG WKH RFFXUUHQFH RI QHJDWLYH OLIH HYHQWV -DFREVRQ 5DQG t +DXVHU f 0RGHUDWRU 9DULDEOHV 'HVSLWH WKH FRQVLVWHQW ILQGLQJ RI VWDWLVWLFDOO\ VLJQLILFDQW FRUUHODWLRQV WKH FRUUHODWLRQ FRHIILFLHQWV DUH RI ORZ PDJQLWXGH W\SLFDOO\ OHVV WKDQ ZKLFK VXJJHVWV WKDW OLIH HYHQWV PHDVXUHV W\SLFDOO\ FDQ RQO\ DFFRXQW IRU b RU VR RI WKH YDULDQFH LQ LOOQHVV PHDVXUHV 5DENLQ t 6WUXHQLQJ f 7R LPSURYH XSRQ WKLV VLWXDWLRQ WKH OLIH HYHQWV PHDVXUHV KDYH EHHQ DGPLQLVWHUHG LQ FRQMXQFWLRQ ZLWK RWKHU PHDVXUHV EHOLHYHG WR DVVHVV PRGHUDWRUV RI OLIH VWUHVV &RQVWLWXWLRQDO SUHGLVSRVLWLRQ .REDVD 0DGGL t &RXULQJWRQ f VRFLDO VXSSRUW 1XFNROOV HW DO 6FKDHIHU &R\QH t /D]DUXV f ORFXV RI FRQWURO 7RYHV 6FKLOO t 5DPDQDLDK f KDUGLQHVV .REDVD f SK\VLFDO ILWQHVV 5RWK t +ROPHV f DQG D WHQGHQF\ WR HQJDJH LQ DURXVLQJ VLWXDWLRQV -RKQVRQ 6DUDVRQ t 6LHJHO f KDYH EHHQ VKRZQ WR PRGHUDWH WKH HIIHFWV RI OLIH VWUHVV r 6RFLDO 6XSSRUW $ IUHTXHQWO\ UHVHDUFKHG PRGHUDWRU YDULDEOH KDV EHHQ VRFLDO VXSSRUW 7KLV UHIHUV WR WKH GHJUHH WR ZKLFK LQGLYLGXDOV KDYH DFFHVV WR VRFLDO UHVRXUFHV 5HVRXUFHV GHSHQG XSRQ WKH UHODWLRQVKLSV SHRSOH KDYH WR VSRXVH IDPLO\ IULHQGV QHLJKERUV FRPPXQLW\ JURXSV DQG VRFLDO LQVWLWXWLRQV -RKQVRQ t 6DUDVRQ f 6RFLDO VXSSRUW KDV

PAGE 12

EHHQ PHDVXUHG LQ D YDULHW\ RI ZD\V IURP VLPSO\ FRQILUPLQJ WKH DYDLODELOLW\ RI D FRQILGDQWH %URZQ %KUROFKDLQ t +DUULV f WR DGPLQLVWHULQJ D LWHP LQYHQWRU\ ZLWK IRXU VXEVFDOHV WR HYDOXDWH HVWHHP VXSSRUW LQVWUXPHQWDO VXSSRUW VRFLDO FRPSDQLRQVKLS DQG LQIRUPDWLRQDO VXSSRUW &RKHQ t +REHUPDQ f ,Q D UHFHQW FRPSUHKHQVLYH UHYLHZ RI RYHU UHVHDUFK DUWLFOHV &RKHQ DQG :LOOV f IRXQG HYLGHQFH FRQVLVWHQW ZLWK ERWK DQ RYHUDOO EHQHILFLDO HIIHFW RI VXSSRUW RQ ZHOOEHLQJ DQG D EXIIHULQJ LQWHUDFWLYHf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t :LOOV f 6RFLDO VXSSRUW ZDV IRXQG WR H[HUW EXIIHULQJ RU SRVLWLYH PDLQ HIIHFWV RQ PRVW RI WKH SV\FKRORJLFDO GLVWUHVV YDULDEOHV UHYLHZHG VXFK DV GHSUHVVLRQ ORQHOLQHVV DQG DQ[LHW\ ,Q IDFW LQ RQH LQYHVWLJDWLRQ /LQ (QVHO 6LPHRQH t .XR f WKH FRQWULEXWLRQ RI VRFLDO VXSSRUW WR SUHGLFWLQJ SV\FKLDWULF V\PSWRPV ZDV PRUH LPSRUWDQW WKDQ VWUHVVIXO OLIH HYHQWV +RZHYHU ILQGLQJV UHJDUGLQJ SK\VLFDO V\PSWRPDWRORJ\ KDYH EHHQ PRUH YDULDEOH $ FOHDU FRQVLVWHQW OLQN EHWZHHQ VRFLDO VXSSRUW DQG GHFUHDVHG PRUWDOLW\ KDV EHHQ GHPRQVWUDWHG %HUNPDQ t 6\PH f EXW VWXGLHV XWLOL]LQJ RWKHU KHDOWK RXWFRPHV DUH OHVV FOHDU 7KH\ DUH

PAGE 13

PL[HG ZKHQ WKH KHDOWK RXWFRPH LV VHOIUHSRUWHG V\PSWRPDWRORJ\ :KHQ PRUH VWULQJHQW KHDOWK PHDVXUHV KDYH EHHQ XVHG VXFK DV FOLQLFDO GLDJQRVLV DQG DOWHUDWLRQV LQ SK\VLRORJLFDO IXQFWLRQLQJ D EHQHILFLDO UROH IRU VRFLDO VXSSRUW KDV QRW DOZD\V EHHQ FRQVLVWHQWO\ GHPRQVWUDWHG &RKHQ t :LOOV f $ IHZ H[DPSOHV ZLOO EH JLYHQ WR LOOXVWUDWH WKH GLYHUVLW\ RI RXWFRPHV UHJDUGLQJ VRFLDO VXSSRUW (YLGHQFH FRQVLVWHQW ZLWK D VWUHVVn EXIIHULQJ K\SRWKHVLV FRPHV IURP DQ HDUO\ VWXG\ RI SUHJQDQW ZRPHQ E\ 1XFNROOV HW DO f 7KRVH H[SHULHQFLQJ KLJK OHYHOV RI VWUHVV DQG ORZ OHYHOV RI VRFLDO VXSSRUW H[SHULHQFHG DOPRVW WKUHH WLPHV DV PDQ\ SUHJQDQF\ DQG ELUWK FRPSOLFDWLRQV DV WKHLU VLPLODUO\ KLJKO\ VWUHVVHG SHHUV ZKR KDG KLJK OHYHOV RI VRFLDO VXSSRUW 6LPLODUO\ DGXOW DVWKPDWLFV LQ D KLJK OLIH VWUHVVORZ VRFLDO VXSSRUW JURXS UHTXLUHG VLJQLILFDQWO\ PRUH PHGLFDWLRQ WR FRQWURO V\PSWRPV WKDQ WKHLU SHHUV ZLWK HLWKHU OHVV VWUHVV RU PRUH VRFLDO VXSSRUW GH $UDXMR HW DO f 7KHVH H[DPSOHV GHPRQVWUDWH D SURWHFWLYH SRVLWLYH UROH IRU VRFLDO VXSSRUW ZLWK UHVSHFW WR FHUWDLQ SK\VLFDO RXWFRPHV ,Q VSLWH RI WKH LQWXLWLYH DSSHDO RI D SRVLWLYH HIIHFW RI VRFLDO VXSSRUW DQG LQ FRQWUDVW WR WKH DERYH VWXGLHV 1RUEHFN DQG 7LOGHQ f IRXQG QR HYLGHQFH IRU HLWKHU D PDLQ HIIHFW RU EXIIHULQJ UROH ZLWK UHVSHFW WR SUHJQDQF\ FRPSOLFDWLRQV 7KH\ GLG KRZHYHU ILQG D SRVLWLYH PDLQ HIIHFW RI HPRWLRQDO VXSSRUW RQ SV\FKRORJLFDO V\PSWRPDWRORJ\ 1RUEHFN t 7LOGHQ f 1HZ =HDODQG UHVHDUFKHUV *UDYHV t *UDYHV f IDLOHG WR ILQG WKH K\SRWKHVL]HG EXIIHULQJ HIIHFW RI VRFLDO VXSSRUW RQ VHOIUHSRUWHG V\PSWRPV RI LOOQHVV LQ WKUHH GLVWLQFW HWKQLF JURXSV ,Q IDFW RI WKH FRUUHODWLRQV GHVLJQDWHG ODUJH

PAGE 14

HQRXJK WR EH RI FOLQLFDO VLJQLILFDQFH DOO ZHUH LQ GLUHFW RSSRVLWLRQ WR WKH EXIIHULQJ K\SRWKHVLV ,W LV VXJJHVWHG WKDW WKH REOLJDWRU\ UHFLSURFLW\ RI GHQVH VRFLDO QHWZRUNV PD\ EH OHVV KHOSIXO XQGHU FRQGLWLRQV UHTXLULQJ DGDSWDWLRQ WR OLIH FKDQJH *UDYHV t *UDYHV f $Q LQWHUHVWLQJ F\FOLFDO SDWWHUQ RI VWUHVV DQG VRFLDO VXSSRUW ZDV GHPRQVWUDWHG E\ &DQDGLDQ LQYHVWLJDWRUV LQ D ORQJLWXGLQDO VWXG\ 0F)DUODQH 1RUPDQ 6WUHLQHU t 5R\ f 7KH\ IRXQG WKDW KHOS IURP VRFLDO VXSSRUW QHWZRUNV UHGXFHG H[SRVXUH WR VWUHVVIXO HYHQWV +RZHYHU LQFUHDVHV LQ VWUHVVIXO HYHQWV OHG WR D UHGXFWLRQ LQ SHUFHLYHG KHS 1R DVVRFLDWLRQ ZDV IRXQG EHWZHHQ VRFLDO VXSSRUW DQG SK\VLFDO V\PSWRP UHSRUWLQJ RU SK\VLFLDQ YLVLWV 0F)DUODQH HW DO f 2EYLRXVO\ WKH UHODWLRQVKLS EHWZHHQ VRFLDO VXSSRUW VWUHVV DQG KHDOWK LV D FRPSOH[ RQH &ULWLFV VXJJHVW WKH XVH RI ORQJLWXGLQDO GHVLJQV &RKHQ t :LOOV 7KRLWV f DQG WKH PHDVXUHPHQW RI VDWLVIDFWLRQ SHUFHLYHG TXDOLW\f ZLWK DYDLODEOH VXSSRUW DV ZHOO DV WKH DPRXQW TXDQWLW\f RI VRFLDO VXSSRUW DYDLODEOH 6DUDVRQ /HYLQH t 6DUDVRQ f WR IXUWKHU XQUDYHO WKH UROH RI VRFLDO VXSSRUW ZLWK UHVSHFW WR PDMRU OLIH HYHQWV DQG KHDOWK VWDWXV /RFXV RI &RQWURO $QRWKHU PHDVXUDEOH YDULDEOH WKDW KDV VKRZQ SURPLVH DV D PRGHUDWRU RI OLIH VWUHVV LV DQ LQGLYLGXDOnV SHUFHSWLRQ RI WKH FRQWUROODELOLW\ RI VLWXDWLRQV 3HRSOH YDU\ LQ WKH GHJUHH WR ZKLFK WKH\ DWWULEXWH WKH UHVSRQVLELOLW\ IRU HYHQWV WR WKHPVHOYHV RU WR IDWH ,W LV UHDVRQDEOH WR DVVXPH WKDW LQGLYLGXDOV ZKR SHUFHLYH WKHPVHOYHV DV KDYLQJ OLWWOH RU QR FRQWURO ZRXOG IHHO PRUH WKUHDWHQHG LQ WKH IDFH RI XQGHVLUDEOH OLIH

PAGE 15

HYHQWV WKDQ WKRVH ZKR IHHO FDSDEOH RI LQIOXHQFLQJ RXWFRPHV $ PRUH LQWHQVH DGYHUVH UHDFWLRQ FRXOG FRQWULEXWH WR D PRUH VHYHUH SK\VLRORJLFDO UHVSRQVH WR OLIH VWUHVV 7KH 5RWWHU f /RFXV RI &RQWURO 6FDOH PHDVXUHV WKH GHJUHH WR ZKLFK LQGLYLGXDOV SHUFHLYH WKHPVHOYHV DV KDYLQJ FRQWURO RYHU QDWXUDOO\ RFFXUULQJ OLIH HYHQWV ,W KDV SURYLGHG D WRRO IRU LQYHVWLJDWLRQ RI SHUFHSWLRQ RI FRQWURO OLIH VWUHVV DQG SV\FKRORJLFDO DQG SK\VLFDO RXWFRPHV ,QGLYLGXDOV DUH FODVVLILHG DV LQWHUQDOV RQ WKH VFDOH LI WKH\ EHOLHYH LQ SHUVRQDO UHVSRQVLELOLW\ DQG H[WHUQDOV LI WKH\ DWWULEXWH WKH FRQWURO RI HYHQWV WR FKDQFH -RKQVRQ DQG 6DUDVRQ f DGPLQLVWHUHG WKH /RFXV RI &RQWURO 6FDOH DORQJ ZLWK D PDMRU OLIH HYHQWV VFDOH DQG PHDVXUHV RI GHSUHVVLRQ DQG DQ[LHW\ WR FROOHJH VWXGHQWV 7KH\ IRXQG QHJDWLYH OLIH FKDQJHV WR EH VLJQLILFDQWO\ UHODWHG WR ERWK WUDLW DQ[LHW\ DQG GHSUHVVLRQ IRU H[WHUQDOV EXW QRW IRU LQWHUQDOV $ VLPLODU UHODWLRQVKLS ZDV IRXQG IRU PDOHV LQ D VWXG\ RI OLIH VWUHVV ORFXV RI FRQWURO DQG KHDOWK VWDWXV DV PHDVXUHG E\ WKH &RUQHOO 0HGLFDO ,QGH[ 7RYHV HW DO f 1HJDWLYH OLIH FKDQJH ZDV UHODWHG WR KHDOWK VWDWXV RQO\ IRU H[WHUQDO PDOHV )HPDOHV GHPRQVWUDWHG D UHODWLRQVKLS EHWZHHQ QHJDWLYH HYHQWV DQG LOOQHVV UHJDUGOHVV RI SHUFHSWLRQ RI FRQWURO 7RYHV HW DO f ,Q IXUWKHU VXSSRUW RI WKH QRWLRQ RI SHUFHLYHG FRQWURO DV D PRGHUDWRU RI OLIH VWUHVV .REDVD f GHPRQVWUDWHG WKDW LQGLYLGXDOV ZLWK DQ LQWHUQDO ORFXV RI FRQWURO PD\ EH SURWHFWHG IURP DGYHUVH SK\VLFDO FRQVHTXHQFHV RI PDMRU OLIH FKDQJH &RPSDUHG WR EXVLQHVV H[HFXWLYHV ZLWK KLJK VWUHVVKLJK LOOQHVV VFRUHV WKRVH ZLWK KLJK VWUHVVORZ LOOQHVV LQGLFDWHG D PRUH LQWHUQDO ORFXV RI FRQWURO ,Q WKLV LQYHVWLJDWLRQ SHUFHSWLRQ RI LQWHUQDO

PAGE 16

FRQWURO HPHUJHG DV RQH FRPSRQHQW RI D FRQVWHOODWLRQ RI IDFWRUV QDPHG KDUGLQHVV .REDVD f 7KH DERYH VWXGLHV DUH DOO UHWURVSHFWLYH LQ QDWXUH DQG LW FRXOG EH DUJXHG WKDW LQGLYLGXDOV ZKR KDYH H[SHULHQFHG UHFHQW LOOQHVV ZRXOG VHH WKHPVHOYHV DV YLFWLPL]HG DQG WKXV WHQG WR UHVSRQG WR WKH ORFXV RI FRQWURO VFDOH DV H[WHUQDOV EHFDXVH RI WKHLU SRRU KHDOWK VWDWXV 7R VWUHQJWKHQ WKH FDXVDO LQIHUHQFH WKDW KDUGLQHVV ZKLFK LQFOXGHV DQ LQWHUQDO ORFXV RI FRQWURO RULHQWDWLRQf .REDVD f UHSOLFDWHG KHU ILQGLQJV LQ D SURVSHFWLYH GHVLJQ VXJJHVWLQJ WKDW DQ LQWHUQDO ORFXV RI FRQWURO LV UHODWHG WR UHVLVWDQFH WR LOOQHVV XQGHU FRQGLWLRQV RI VWUHVV 0DQXFN +LQULFKVHQ DQG 5RVV f IROORZHG FROOHJH VWXGHQWV IRU PRQWKV DIWHU DVVHVVLQJ ORFXV RI FRQWURO RULHQWDWLRQ DQG OLIH VWUHVV $OWKRXJK WKH\ IRXQG QR GLIIHUHQFH LQ LOOQHVVUHODWHG WUHDWPHQWVHHNLQJ EHKDYLRU EHWZHHQ LQWHUQDOV DQG H[WHUQDOV XQGHU KLJK VWUHVV FRQGLWLRQV PRUH ORZVWUHVVHG H[WHUQDOV VRXJKW WUHDWPHQW WKDQ ORZVWUHVVHG LQWHUQDOV 0DQXFN HW DO f 3DVVHU DQG 6HHVH f KRZHYHU IDLOHG WR SURVSHFWLYHO\ GHPRQVWUDWH D VLJQLILFDQW PRGHUDWLQJ HIIHFW RI SHUFHSWLRQ RI FRQWURO ZLWK DWKOHWLF LQMXU\ DV WKH GHSHQGHQW YDULDEOH 7KHVH ILQGLQJV VXJJHVW WKDW ORFXV RI FRQWURO PHULWV IXUWKHU LQYHVWLJDWLRQ DV D PRGHUDWRU RI WKH HIIHFWV RI PDMRU OLIH VWUHVV $URXVDO 6HHNLQJ ,QGLYLGXDOV GLIIHU LQ WKH GHJUHH WR ZKLFK WKH\ VHHN RXW RU DWWHPSW WR DYRLG DURXVLQJ VLWXDWLRQV ,QYHQWRULHV KDYH EHHQ GHYLVHG WR HYDOXDWH WKLV WHQGHQF\ HQDEOLQJ UHVHDUFKHUV WR LQYHVWLJDWH WKLV

PAGE 17

FKDUDFWHULVWLF DV D PRGHUDWRU RI OLIH VWUHVV 0HKUDELDQ 0HKUDELDQ t 5XVVHOO =XFNHUPDQ f ,W LV K\SRWKHVL]HG WKDW WKRVH ZLWK D ORZ RSWLPDO OHYHO RI DURXVDO ZKR WHQG WR DYRLG VWLPXODWLRQ ORZ DURXVDO VHHNHUVf ZRXOG EH PRUH OLNHO\ WR UHVSRQG QHJDWLYHO\ WR XQGHVLUDEOH OLIH HYHQWV WKDQ WKRVH ZKR KDYH D KLJK RSWLPDO OHYHO RI DURXVDO DQG WKXV VHHN RXW VWLPXODWLRQ KLJK DURXVDO VHHNHUVf +LJK DURXVDO VHHNHUV DUH SUHVXPHG WR EH EHWWHU DEOH WR GHDO ZLWK WKH LQFUHDVHG DURXVDO EURXJKW DERXW E\ WKH H[SHULHQFLQJ RI OLIH FKDQJHV -RKQVRQ t 6DUDVRQ f 7KLV SUHGLFWLRQ KDV EHHQ DGGUHVVHG E\ D IHZ UHVHDUFKHUV 6PLWK -RKQVRQ DQG 6DUDVRQ f IRXQG QHJDWLYH OLIH FKDQJH WR EH VLJQLILFDQWO\ UHODWHG WR GLVFRPIRUW VFRUHV RQ WKH 3V\FKRORJLFDO 6FUHHQLQJ ,QYHQWRU\ IRU RQO\ WKH FROOHJH VWXGHQWV ORZ RQ D VHQVDWLRQ VHHNLQJ PHDVXUH ([WHQGLQJ WKHVH ILQGLQJV XVLQJ D GLIIHUHQW PHDVXUH RI DURXVDO VHHNLQJ DQG PHDVXUHV RI DQ[LHW\ GHSUHVVLRQ DQG KRVWLOLW\ -RKQVRQ HW DO f IRXQG QHJDWLYH FKDQJH WR EH UHODWHG WR DQ[LHW\ DQG KRVWLOLW\ RQO\ IRU ORZ DURXVDO VHHNLQJ VXEMHFWV $W OHDVW RQH UHVHDUFKHU &RKHQ f KDV IDLOHG WR UHSOLFDWH WKH DERYH UHVXOWV 7KHVH SUHOLPLQDU\ ILQGLQJV ZLWK UHJDUG WR SV\FKRORJLFDO RXWFRPHV VXJJHVW WKDW D PRGHUDWLQJ HIIHFW RI DURXVDO VHHNLQJ WHQGHQF\ RQ OLIH VWUHVV ZLWK UHJDUG WR FHUWDLQ SK\VLFDO RXWFRPHV PD\ DOVR EH ZRUWK LQYHVWLJDWLRQ ,QGLYLGXDOV ORZ RQ WKH DURXVDOVHHNLQJ GLPHQVLRQ PD\ EH PXFK PRUH OLNHO\ WR H[SHULHQFH SK\VLRORJLFDO UHDFWLYLW\ LQ WKH IDFH RI OLIH VWUHVV WKDQ RWKHUV ZKLFK FRXOG DGYHUVHO\ DIIHFW JHQHUDO KHDO WK

PAGE 18

,Q VXPPDU\ DV D PHDQV RI LQYHVWLJDWLQJ WKH YDULDELOLW\ LQ LQGLYLGXDO VXVFHSWLELOLW\ WR KHDOWK VWDWXV FKDQJHV LQ WKH IDFH RI XQGHVLUDEOH OLIH VWUHVV WKH LQYHVWLJDWLRQ RI WKH UROHV RI PRGHUDWRU YDULDEOHV DSSHDUV WR EH ZDUUDQWHG 7KH SRWHQWLDO LQIOXHQFH RI VRFLDO VXSSRUW ORFXV RI FRQWURO DQG DURXVDO VHHNLQJ WHQGHQF\ DUH RI SDUWLFXODU LQWHUHVW EDVHG RQ ILQGLQJV WR GDWH :LWK WKH H[FHSWLRQ RI WKH .REDVD LQYHVWLJDWLRQV .REDVD .REDVD HW DO f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t 6DUDVRQ f ,Q WKH QH[W VHFWLRQ VRPH RI WKH PHWKRGRORJLFDO VKRUWFRPLQJV RI SUHVHQW UHVHDUFK ZLOO EH GLVFXVVHG $GGUHVVLQJ WKHVH LVVXHV ZLOO HQDEOH XV WR EH PRUH VSHFLILF DERXW RXU UHVHDUFK TXHVWLRQV DQG WR IXUWKHU FODULI\ WKH QDWXUH RI WKH VWUHVVL QHVV UHODWLRQVKLS

PAGE 19

0HWKRGRORJLFDO ,VVXHV 0HWKRGRORJLFDO LVVXHV OLPLW LQWHUSUHWDWLRQV DQG ILQGLQJV LQ WKLV DUHD RI UHVHDUFK 7KHVH LQFOXGH UHWURVSHFWLYH ELDV LQDGHTXDWH LQGLFHV RI LOOQHVV IDLOXUH WR GLVFULPLQDWH EHWZHHQ GHVLUDEOH DQG XQGHVLUDEOH HYHQWV DQG IDLOXUH WR DFFRXQW IRU PLQRU VWUHVVIXO HYHQWV 5HWURVSHFWLYH 'HVLJQ 7KH FRPPRQO\ XWLOL]HG UHWURVSHFWLYH GHVLJQ UHOLHV RQ UHFDOO RI HYHQWV DQG LOOQHVVHV )LQGLQJV EDVHG RQ PHPRU\ PD\ UHIOHFW LQGLYLGXDOVn H[SHFWDWLRQV UHJDUGLQJ VWUHVV DQG LOOQHVV WKHLU QHHG WR MXVWLI\ LOOQHVV 5DENLQ t 6WUXHQLQJ f RU WKHLU WHQGHQF\ WR UHSRUW QHJDWLYHV GXH WR IHHOLQJV RI GHSUHVVLRQ -RKQVRQ t 6DUDVRQ f 5HWURVSHFWLYH UHSRUWLQJ LV DOVR VXEMHFW WR IRUJHWWLQJ DV GHPRQVWUDWHG E\ 0RQURH Df +H LQYHVWLJDWHG PHPRU\ RI PDMRU OLIH HYHQWV XWLOL]LQJ SURVSHFWLYHO\ FROOHFWHG LQIRUPDWLRQ DV D EDVHOLQH WR LQGLUHFWO\ HVWLPDWH WKH PRVW UHFHQW UHWURVSHFWLYHO\ UHSRUWHG IRXU PRQWK SHULRG 8QGHUUHSRUWLQJ RI HYHQWV ZDV HVWLPDWHG LQ WKLV ZD\ WR EH DV KLJK DV ZLWK GHVLUDEOH HYHQWV UHODWLYHO\ PRUH VXVFHSWLEOH WR GLVWRUWLRQ ZLWK SDVVLQJ WLPH WKDQ XQGHVLUDEOH HYHQWV 0RQURH Df )XQFK DQG 0DUVKDOO f H[DPLQHG WKH UDWH RI IDOORII LQ HYHQW UHSRUWLQJ RYHU D UHWURVSHFWLYH PRQWK SHULRG 8VLQJ HVWLPDWHV EDVHG RQ DQG PRQWK UHFDOO SHULRGV WKH\ FRQFOXGHG WKDW IDOORII LV PRVW UDSLG LQ WKH ILUVW PRQWKV DSSUR[LPDWHO\ b SHU PRQWK ,W WKHQ WHQGV WR OHYHO RII )DOORII ZDV UHODWHG WR W\SH RI HYHQW LQ WKHLU UHVHDUFK WKH PRUH VDOLHQW HYHQWV VKRZLQJ WKH OHDVW IDOORII 5HVSRQGHQW YDULDEOHV VXFK DV LQFRPH HGXFDWLRQ DQG PDUULDJH ZHUH

PAGE 20

UHODWHG WR WKH WHQGHQF\ WR UHPHPEHU GLVUXSWLYH HYHQWV 7KDW LV VXEMHFWV ZLWK IHZHU UHVRXUFHV ZHUH PRVW OLNHO\ WR XQGHUUHSRUW VWUHVVIXO HYHQWV )XQFK t 0DUVKDOO f ,Q D UHYLHZ RI VHYHUDO DUWLFOHV ZKLFK DGGUHVV WKH LVVXH RI HYHQW XQGHUUHSRUWLQJ 3D\NHO f IRXQG FRQVLVWHQF\ LQ HVWLPDWHV RI b ORVV SHU PRQWK LQ UHWURVSHFWLYHO\ VHOIUHSRUWHG OLIH HYHQWV ZLWK D ORZHU UDWH bf IRU HYHQW UHSRUWLQJ E\ LQWHUYLHZ 7R DYRLG UHWURVSHFWLYH ELDV GXH WR IRUJHWWLQJ IUHTXHQW VKRUWWLPH LQWHUYDOV IRU UHSRUWLQJ KDYH EHHQ UHFRPPHQGHG &OHDU\ 0RQURH Df ZLWK LQWHUYLHZ WHFKQLTXHV XVHG ZKHQHYHU SRVVLEOH 3D\NHO f 3URVSHFWLYH VWXGLHV DYRLG WKLV NLQG RI GLVWRUWLRQ -HQNLQV +XUVW DQG 5RVH f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t 1LHGHUPDQ f ,Q DQRWKHU H[DPSOH ZLGRZHUV ZHUH IROORZHG IRU \HDUV WR FRQFOXGH WKDW WKHUH LV D b LQFUHDVH LQ PRUWDOLW\ ZLWKLQ WKH ILUVW PRQWKV RI EHUHDYHPHQW 3DUNHV %HQMDPLQ t )LW]JHUDOG f 3URVSHFWLYH LQYHVWLJDWLRQV RI WKH HIIHFWV RI OLIH VWUHVV RQ WKH FOLQLFDO FRXUVH DQG RXWFRPH RI GLVHDVH DUH PRUH HDVLO\ DQG HFRQRPLFDOO\

PAGE 21

DFFRPSOLVKHG 7KHVH LQYROYH SURVSHFWLYHO\ IROORZLQJ DOUHDG\ LGHQWLILHG SDWLHQWV ZLWK UHVSHFW WR OLIH VWUHVV DORQJ ZLWK IOXFWXDWLRQV LQ WKHLU KHDOWK VWDWXV UHJDUGLQJ WKH LOOQHVV LQ TXHVWLRQ 7KH GDWD IURP DOO VXEMHFWV UHPDLQ LPSRUWDQW IRU DQDO\VLV QRW MXVW WKRVH ZKR KDSSHQ WR GLH RU EHFRPH LQIHFWHG ZLWK D SDUWLFXODU GLVHDVH 0HDVXUHV RI ,OOQHVV $QRWKHU ZHDNQHVV RI UHVHDUFK LQ WKLV DUHD KDV EHHQ WKH IDLOXUH WR HPSOR\ VWULFW PHDVXUHV RI LOOQHVV -RKQVRQ t 6DUDVRQ f 6HOI UHSRUWV RI V\PSWRPV WKH VHHNLQJ RI PHGLFDO WUHDWPHQW DQG PHGLFDWLRQ QHHGV PD\ QRW UHIOHFW WKH H[LVWHQFH RU VHYHULW\ RI D SK\VLFDO GLVRUGHU $GGLWLRQDOO\ KLJK OLIH VWUHVV PD\ LQFUHDVH WKH WHQGHQF\ WR VHHN KHOS IRU PLQRU LOOQHVV 0HFKDQLF f RU ORZHU DQ LQGLYLGXDOnV WKUHVKROG RI FRPSODLQW DQG WHQGHQF\ WR DGRSW WKH VLFN UROH 0LQWHU t .LPEDOO f +DUQH\ DQG %ULJKDP f H[SHULPHQWDOO\ GHPRQVWUDWHG WKDW DQ LQGLYLGXDOnV WHQGHQF\ WR WROHUDWH GLVFRPIRUW FROG SUHVVRU ORXG QRLVHf LV UHODWHG WR WKH DPRXQW RI UHFHQW OLIH FKDQJH H[SHULHQFHG 7KLV VXJJHVWV WKDW KLJK OLIH FKDQJH LQGLYLGXDOV PD\ EH PRUH OLNHO\ WR VHHN DLG WKDQ RWKHUV IRU LGHQWLFDO V\PSWRPV +DUQH\ t W %ULJKDP f 2QH ZD\ WR DYRLG WKHVH FRQIRXQGLQJ IDFWRUV LV WR HPSOR\ PRUH ULJLG PHDVXUHV RI LOOQHVV 5DWKHU WKDQ UHOLDQFH RQ LOOQHVV EHKDYLRU PHGLFDO UHFRUGV RU VHOIUHSRUW PRUH REMHFWLYH LQGLFHV DUH VXJJHVWHG VXFK DV [UD\ DQG RWKHU FOLQLFDO DQG ODERUDWRU\ GDWD ZKLFK SURYLGH ELRORJLFDO HYLGHQFH RI WKH SUHVHQFH RU VHYHULW\ RI GLVHDVH 0LQWHU t .LPEDOO f

PAGE 22

&RQIRXQGLQJ RI (YHQWV DQG 2XWFRPH 0HDVXUHV $ FRQFHUQ RI PDQ\ LQ WKLV DUHD RI UHVHDUFK LQ WKH SRVVLEOH FRQIRXQGLQJ RI LQGHSHQGHQW DQG GHSHQGHQW YDULDEOHV LQ OLIH VWUHVV PHDVXUHV -RKQVRQ t 6DUDVRQ 0RQURH Ef 0DQ\ LWHPV ZKLFK DUH FRQVLGHUHG WR EH VWUHVVIXO HYHQWV PD\ DOVR EH FRQVHTXHQFHV RI LOOQHVV HJ VH[XDO SUREOHPV PDMRU SHUVRQDO LQMXU\ FKDQJHV LQ HDWLQJ RU VOHHSLQJ KDELWVf $Q DUJXPHQW FDQ EH PDGH WKDW VXFK LWHPV VKRXOG EH UHPRYHG RU DQDO\]HG VHSDUDWHO\ IURP WKH GRPDLQ RI PDMRU OLIH HYHQWV WR DYRLG DUWLILFLDOO\ LQIODWLQJ WKH VWUHVV VFRUHV RI LQGLYLGXDOV ZLWK LOOQHVV 'RKUHQZHQG t 'RKUHQZHQG f +RZHYHU VXFK H[SHULHQFHV PD\ WKHPVHOYHV UHSUHVHQW VLJQLILFDQW VRXUFHV RI VWUHVV 7R LJQRUH WKHP LQ D FRQVLGHUDWLRQ RI WKH WRWDO VWUHVV RI DQ LQGLYLGXDO PD\ EH WR QHJOHFW LPSRUWDQW LQIRUPDWLRQ DQG UHVXOW LQ D ORVV RI VHQVLWLYLW\ LQ WKH OLIH VWUHVV PHDVXUH ,Q DQ LQYHVWLJDWLRQ RI HWLRORJLFDO IDFWRUV DVVRFLDWHG ZLWK D SDUWLFXODU GLVHDVH D GLYLVLRQ RI LWHPV LQWR WKUHH VXEJURXSV IRU DQDO\VLV LV UHFRPPHQGHG Df HYHQWV WKDW PD\ EH FRQIRXQGHG ZLWK SV\FKLDWULF FRQGLWLRQ Ef HYHQWV FRQVLVWLQJ RI SK\VLFDO LOOQHVV DQG LQMXU\ DQG Ff HYHQWV ZKLFK DUH LQGHSHQGHQW RI HLWKHU DQ LQGLYLGXDOnV KHDOWK RU SV\FKLDWULF FRQGLWLRQ 'RKUHQZHQG t 'RKUHQZHQG 0RQURH Ef :KHQ WKH PDMRU SXUSRVH RI WKH LQYHVWLJDWLRQ LV WR SUHGLFW LOOQHVV IURP VWUHVV OHYHOV LW LV UHDVRQDEOH WR LQFOXGH DOO LWHPV DV UHIOHFWLYH RI WKH WRWDO VWUHVV H[SHULHQFHG -RKQVRQ t 6DUDVRQ f 7KLV ZRXOG EH SDUWLFXODUO\ DSSURSULDWH ZKHQ WKH LOOQHVV GHSHQGHQW YDULDEOH LV VSHFLILF DQG REMHFWLYH HJ ODERUDWRU\ GRFXPHQWDWLRQ RI WKH SUHVHQFH

PAGE 23

RI D GLVHDVH RUJDQLVPf &RQIRXQGLQJ SUREOHPV DUH PRUH OLNHO\ ZKHQ JHQHUDO VXEMHFWLYH PHDVXUHV RI LOOQHVV VHUYH DV WKH GHSHQGHQW YDULDEOH HJ UHSRUWHG V\PSWRPV WUHDWPHQWVHHNLQJ EHKDYLRUf 9DOXH DQG ,PSDFW RI (YHQWV 2WKHU PHWKRGRORJLFDO FRQVLGHUDWLRQV LQFOXGH WKH YDOXH QHJDWLYH RU SRVLWLYHf DQG LPSDFW ZHLJKWLQJf RI HYHQWV 8QWLO UHFHQWO\ LQYHVWLJDWRUV KDYH IDLOHG WR GLVFULPLQDWH EHWZHHQ GHVLUDEOH DQG XQGHVLUDEOH HYHQWV ,Q WKRVH LQYHVWLJDWLRQV LQ ZKLFK D GLVWLQFWLRQ KDV EHHQ PDGH QHJDWLYH FKDQJH KDV SURYHQ WR EH PRUH SUHGLFWLYH RI SHUVRQDO PDODGMXVWPHQW 0XHOOHU (GZDUGV t
PAGE 24

XQGHUVWDQGLQJ SDUWLFXODU LQGLYLGXDOV ,W VLPSO\ DFNQRZOHGJHV WKDW VXFK UHILQHPHQWV DUH XQQHFHVVDU\ DQG UHGXQGDQW IRU JURXS UHVHDUFK 0LQRU (YHQWV 7KH ILQDO PHWKRGRORJLFDO LVVXH UHODWHV WR WKH FRQVLGHUDWLRQ RI PLQRU DV ZHOO DV PDMRU OLIH HYHQWV %HFDXVH PDMRU OLIH HYHQWV SHU VH GR QRW UHSUHVHQW DOO VWUHVVIXO HYHQWV WKH +DVVOHV 6FDOH +6f KDV EHHQ GHYHORSHG .DQQHU &R\QH 6FKDHIHU t /D]DUXV f ,Q FRQWUDVW WR PDMRU OLIH HYHQWV VXFK DV GLYRUFH MRE ORVV DQG SUHJQDQF\ WKH +6 GHDOV ZLWK PLQRU VWUHVVHV DQQR\LQJ SUDFWLFDO SUREOHPV VXFK DV PDNLQJ PLVWDNHV EHLQJ ERUHG DQG KDYLQJ WR ZDLW $OUHDG\ WKHUH LV VRPH VXJJHVWLRQ WKDW KDVVOH VFRUHV PD\ EH PRUH VWURQJO\ UHODWHG WR D JHQHUDO KHDOWK PHDVXUH WKDQ OLIH HYHQWV VFRUHV 'H/RQJLV &R\QH 'DNRI )RONPDQ t /D]DUXV f ,Q RUGHU WR DFFRXQW IRU D JUHDWHU SURSRUWLRQ RI WKH YDULDQFH LQ WKH UHODWLRQVKLS EHWZHHQ VWUHVV DQG LOOQHVV WKH SUHVHQW VWXG\ ZDV GHVLJQHG WR DGGUHVV VRPH RI WKH H[LVWLQJ PHWKRGRORJLFDO OLPLWDWLRQV 7R DYRLG UHWURVSHFWLYH ELDV D SURVSHFWLYH UHVHDUFK GHVLJQ ZLWK PRQWKO\ FRQFXUUHQW DVVHVVPHQWV RI VWUHVVRUV ZDV SODQQHG 7KH /LIH ([SHULHQFHV 6XUYH\ /(6f ZDV FKRVHQ WR PHDVXUH PDMRU OLIH VWUHVV EHFDXVH LW GLVWLQJXLVKHV EHWZHHQ QHJDWLYH XQGHVLUDEOHf DQG SRVLWLYH GHVLUDEOHf HYHQWV 6DUDVRQ -RKQVRQ t 6LHJHO f 7KH +DVVOHV 6FDOH +6f ZDV XWLOL]HG WR HYDOXDWH PLQRU VWUHVVIXO HYHQWV .DQQHU HW DO f $GGLWLRQDOO\ WKUHH K\SRWKHVL]HG PRGHUDWRU YDULDEOHV ZHUH PHDVXUHG WR HYDOXDWH WKHLU HIIHFWLYHQHVV LQ FRPELQDWLRQ LQ

PAGE 25

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f W\SHV DQG DUH WZR RI VHYHUDO KHUSHV YLUXVHV ZKLFK DIIHFW KXPDQV DQG VKDUH D XQLTXH FDSDFLW\ WR SURGXFH OLIHORQJ ODWHQW LQIHFWLRQ GHVSLWH KRVW GHYHORSPHQW RI DQWLERG\ DQG FHOOXODU LPPXQLW\ 6HULRXV FRPSOLFDWLRQV FDQ RFFXU LQFOXGLQJ RFXODU LQIHFWLRQ HQFHSKDOLWLV DQG QHRQDWDO LQIHFWLRQ )RUWXQDWHO\ WKHVH DUH UDUH DQG FDQ XVXDOO\ EH SUHYHQWHG 7KHUH DOVR DSSHDUV WR EH DQ DVVRFLDWLRQ EHWZHHQ JHQLWDO KHUSHV DQG FHUYLFDO FDQFHU EXW VR IDU GLUHFW HYLGHQFH RI D FDXVDO OLQN LV ODFNLQJ 5DQG f 1HLWKHU VHULRXV FRPSOLFDWLRQV QRU VLJQLILFDQW SK\VLFDO FRQVHTXHQFHV DUH UHVSRQVLEOH IRU WKH ZLGHVSUHDG IHDU DQG DWWHQWLRQ HOLFLWHG E\ WKH GLVHDVH 7KHUH LV HYLGHQFH WKDW WKH LQFLGHQFH RI JHQLWDO KHUSHV LV ULVLQJ UDSLGO\ &RQVXOWDWLRQV WR FOLQLFLDQV IRU

PAGE 26

JHQLWDO KHUSHV KDYH LQFUHDVHG DOPRVW QLQHIROG LQ WKH SDVW \HDUV 0HUW] t &RUH\ f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f LQIHFWLRQ WKLV SURFHVV SURGXFHV PXOWLSOH SDLQIXO EOLVWHUV RU XOFHUDWLYH OHVLRQV DFFRPSDQLHG E\ IHYHU VZROOHQ O\PSK QRGHV DQG PDODLVH 7KH OHVLRQV ODVW DQ DYHUDJH RI GD\V 0HUW] t &RUH\ f :LWKLQ KRXUV RI LQLWLDO H[SRVXUH WKH YLUXV UHWUHDWV YLD QHUYH ILEHUV WR D VDIH KLGLQJ SODFH LQ WKH WULJHPLQDO JDQJOLD RUDO KHUSHVf RU WKH VDFUDO JDQJOLD JHQLWDO KHUSHVf ZKLFK LV ORFDWHG QHDU WKH VSLQDO FRUG ,W LV WKLV FDSDFLW\ RI WKH YLUXV WR UHVLGH ODWHQWO\ ZLWKRXW GHVWUR\LQJ WKH JDQJOLRQ DQG WR SHULRGLFDOO\ UHDFWLYDWH WR FDXVH UHFXUUHQW OHVLRQV ZKLFK IUXVWUDWHV SDWLHQWV GRFWRUV DQG UHVHDUFKHUV 5HFXUUHQFHV $OPRVW DOO SDWLHQWV H[SHULHQFH UHFXUUHQFHV IURP RQH LQ D OLIHWLPH WR WZR RU WKUHH D PRQWK 7KH ULVN RI UHFXUUHQFH ZKHQ WKH LQIHFWLRXV DJHQW LV +69 LV KLJKHU WKDQ ZLWK +69 UHJDUGOHVV RI WKH

PAGE 27

VLWH RI WKH LQIHFWLRQ 2WKHUZLVH WKH FOLQLFDO PDQLIHVWDWLRQ RI +69 DQG +69 LV LGHQWLFDO 0HUW] t &RUH\ f $ GLYHUVH JURXS RI IDFWRUV DUH DVVRFLDWHG ZLWK UHDFWLYDWLRQ RI +69 LQFOXGLQJ IHYHU .HGGLH 5HHYHV t (SVWHLQ f ORFDO LUULWDWLRQ RU WLVVXH LQMXU\ %O\WK t +LOO :LFNHWW f ODFN RI VOHHS RYHUH[HUWLRQ DQG HPRWLRQDO VWUHVV %LHUPDQ f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t &RUH\ f 3DWLHQWV ILQG WKH SHULRGLF LPSRVHG DEVWLQHQFH IURP VH[XDO DFWLYLW\ WR EH GLVUXSWLYH DQG WR UHVXOW LQ GHFUHDVHG VH[XDO VSRQWDQHLW\ )HPDOH SDWLHQWV KDYH FRQFHUQV DERXW WKHLU UHSURGXFWLYH FDSDELOLW\ 6LQJOH LQGLYLGXDOV PXVW GHDO ZLWK GLVFORVXUH WR SRWHQWLDO SDUWQHUV FDXVLQJ HPEDUUDVVPHQW VKDPH DQG IHDU RI UHMHFWLRQ /XE\ t .OLQJH f 8QIRUWXQDWHO\ EHFDXVH RI LWV VH[XDO QDWXUH ELDVHG UHSRUWLQJ UHJDUGLQJ JHQLWDO KHUSHV LQ WKH PHGLD KDV FRQWULEXWHG WR SRSXODU SDQLF DERXW WKH GLVHDVH VWLJPDWL]LQJ SDWLHQWV DQG H[DFHUEDWLQJ WKH HPRWLRQDO WXUPRLO LQYROYHG %LHUPDQ f

PAGE 28

3UHVHQW /DFN RI &XUH $W WKH SUHVHQW WLPH QR FXUH LV DYDLODEOH IRU WKH GLVHDVH ,W KDV UHFHQWO\ EHHQ GHPRQVWUDWHG WKDW RUDO DGPLQLVWUDWLRQ RI GDLO\ GRVHV RI DF\FORYLU VXSSUHVVHV UHDFWLYDWLRQ RI WKH YLUXV LQ SDWLHQWV ZLWK IUHTXHQWO\ UHFXUULQJ JHQLWDO KHUSHV 'RXJODV HW DO 6WUDXV HW DO f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b RI XOFHUDWLYH OHVLRQV LQ

PAGE 29

SULPDU\ JHQLWDO KHUSHV HSLVRGHV DQG } RI XOFHUDWLYH OHVLRQV LQ UHFXUUHQW HSLVRGHV 7KH GHFUHDVH LQ SHUFHQW YLUDO FRQILUPDWLRQ LQ UHFXUUHQW HSLVRGHV DV RSSRVHG WR SULPDU\ FDQ EH DWWULEXWHG WR D VKRUWHU GXUDWLRQ RI YLUDO VKHGGLQJ LQ WKH IRUPHU &RUH\ t +ROPHV f 7KH IDFW WKDW REMHFWLYH HYLGHQFH RI WKH SUHVHQFH RI WKH GLVHDVH LV SRVVLEOH DOVR FRQWULEXWHG WR WKH FKRLFH WR VWXG\ WKLV SDUWLFXODU JURXS RI LQGLYLGXDOV LQ D VWUHVVLOOQHVV SDUDGLJP (QGRJHQRXV 3DWKRJHQ $ ILQDO IDFWRU UHJDUGLQJ UHFXUUHQW JHQLWDO KHUSHV LV WKDW H[WUDQHRXV H[SRVXUH WR DQ XQNQRZQ GRVDJH RI DQ H[RJHQRXV SDWKRJHQ ZDV QRW DQ LVVXH 6WUHVV ZRXOG EH PRVW LPSRUWDQW DV D FRQWULEXWLQJ IDFWRU LQ WKH HWLRORJ\ RI D GLVHDVH ZKHQ WKH KRVWPLFURRUJDQLVP UHODWLRQVKLS LV LQ GHOLFDWH EDODQFH 3ODXW t )ULHGPDQ f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

PAGE 30

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f UHVHDUFK LQFOXGLQJ HLWKHU +69 RU +69 ZLOO EH UHYLHZHG $QHFGRWDO 5HSRUWV 6FKQHFN f UHSRUWHG WKH FDVH RI D PDOH ZKR FRXOG SUHGLFW +69 UHFXUUHQFHV ZKLFK RFFXUUHG ZLWKLQ KRXUV RI HPRWLRQDO VWUHVV 7KHVH HSLVRGHV ZHUH PRVW OLNHO\ WR RFFXU ZKHQ DSSURSULDWH RXWOHWV IRU KRVWLOH IHHOLQJV ZHUH XQDYDLODEOH $FFRUGLQJ WR WKH DFFRXQW LPSURYHG PDQDJHPHQW RI KRVWLOLW\ UHVXOWHG LQ GHFUHDVHG UHFXUUHQFHV %ODQN DQG %URG\ f SURYLGHG SV\FKRDQDO\VLV IRU SDWLHQWV ZLWK UHFXUUHQW RUDO KHUSHV 3DWLHQWV ZHUH VHHQ ZHHNO\ IRU WR KRXUO\ VHVVLRQV 1LQH RI WKH WHQ SDWLHQWV ZHUH GHVFULEHG DV SDVVLYH DQ[LRXV WR SOHDVH GHSHQGHQW LPPDWXUH DQG K\VWHULFDO ,Q WZR FDVHV

PAGE 31

FDQFHOODWLRQ RI DQ DSSRLQWPHQW E\ WKH WKHUDSLVW DQG WKH UHVXOWDQW IHHOLQJ RI UHMHFWLRQ LV VDLG WR KDYH SUHFLSLWDWHG D KHUSHV UHFXUUHQFH )URP WKHLU H[SHULHQFHV ZLWK WKLV JURXS RI SDWLHQWV DXWKRUV FODLP VXSSRUW IRU WKH XVH RI SV\FKRWKHUDS\ LQ GHFUHDVLQJ WKH IUHTXHQF\ RI UHFXUUHQFHV %ODQN t %URG\ f 7KHUH KDYH EHHQ UHSRUWV RI DQ DVVRFLDWLRQ EHWZHHQ K\SQRWLF VXJJHVWLRQ DQG UHFXUUHQW KHUSHV HSLVRGHV +HUOLJ DQG +RII FLWHG LQ -DQLFNL f VXFFHVVIXOO\ LQGXFHG RUDO KHUSHV UHFXUUHQFHV LQ WKUHH IHPDOH SDWLHQWV E\ UHPLQGLQJ WKHP RI XQSOHDVDQW HPRWLRQDO VLWXDWLRQV DQG LWFKLQJ VHQVDWLRQV ZKLOH XQGHU K\SQRVLV 5HFXUUHQW OHVLRQV DSSHDUHG LQ RU GD\V LQ DOO WKUHH VXEMHFWV 8OOPDQ f VLPLODUO\ LQGXFHG DQ RUDO KHUSHV HSLVRGH LQ D PDOH SDWLHQW E\ JLYLQJ K\SQRWLF VXJJHVWLRQV WKDW KHUSHV EOLVWHUV ZRXOG IRUP DQG WKDW KH DSSHDUHG GHELOLWDWHG :LWKLQ KRXUV WKH PDQ KDG EOLVWHUV RQ KLV ORZHU ULJKW OLS ZKLFK DSSHDUHG WR EH RUDO KHUSHV +\SQRVLV KDV DOVR EHHQ UHSRUWHG WR KDYH D SRVLWLYH HIIHFW RQ WKH UHFXUUHQFH UDWH RI +69 0F'RZHOO f 0DQ\ VXSSRVHG FXUHV RI KHUSHV KDYH EHHQ DWWULEXWHG WR SODFHER HIIHFWV RU SRVLWLYH H[SHFWDQF\ 2QH LQYHVWLJDWRU UHSRUWV IDYRUDEOH UHDFWLRQV LQ b RI KHUSHV SDWLHQWV ZKR UHFHLYHG LQMHFWLRQV RI VWHULO ZDWHU .HUQ f +DPLOWRQ f UHSRUWV HYHQ PRUH GUDPDWLF SODFHER UHVXOWV ,Q D GRXEOHEOLQG SODFHERFRQWUROOHG WULDO RI HWKHU DV D WUHDWPHQW IRU KHUSHV b RI WKH SDWLHQWV RQ SODFHER H[SHULHQFHG LPSURYHPHQW RI KHUSHV V\PSWRPDWRORJ\ +DPLOWRQ S f $QHFGRWDO UHSRUWV RI DQ DVVRFLDWLRQ EHWZHHQ K\SQRVLV RU H[SHFWDQF\ DQG HSLVRGHV RI UHFXUUHQW KHUSHV DUH VXJJHVWLYH EXW

PAGE 32

LQFRQFOXVLYH )DLOXUHV WR LQGXFH KHUSHV EOLVWHUV WKURXJK VXJJHVWLRQ DUH XQUHSRUWHG +RZHYHU WKH H[LVWHQFH RI D OLQN EHWZHHQ HPRWLRQDO RU FRJQLWLYH IDFWRUV DQG UHFXUUHQW KHUSHV DV FDQ EH K\SRWKHVL]HG IURP WKHVH UHSRUWV LV ZRUWK\ RI PRUH V\VWHPDWLF LQYHVWLJDWLRQ 6XUYH\ 5HVHDUFK 5HVXOWV RI HSLGHPLRORJLFDO VXUYH\ UHVHDUFK DOVR VXJJHVW D OLQN EHWZHHQ HPRWLRQDO IDFWRUV DQG UHFXUUHQFHV ,Q D QDWLRQDO VXUYH\ RI LQGLYLGXDOV ZLWK JHQLWDO KHUSHV b RI WKH UHVSRQGHQWV DFNQRZOHGJHG VWUHVV DV D IDFWRU LQ UHFXUUHQFHV +HOS PHPEHUVKLS +69 VXUYH\ f 7KH VXEMHFWV DWWULEXWHG WKH IROORZLQJ LOO HIIHFWV WR KDYLQJ FRQWUDFWHG KHUSHV SHULRGLF GHSUHVVLRQV bf VHQVH RI LVRODWLRQ bf FRQVFLRXV DYRLGDQFH RI LQWLPDF\ bf FHVVDWLRQ RI VH[XDO DFWLYLW\ bf GLPLQLVKHG VH[ GULYH DQGRU LPSRWHQFH bf VXLFLGDO IHHOLQJV bf GLVVROXWLRQ RI D ORQJ WHUP PDUULDJH RU UHODWLRQVKLS bf UHMHFWLRQ E\ D SRWHQWLDO VH[ SDUWQHU bf DQG ORVV RI VHOI HVWHHP bf +HUSHV LV FLWHG DV ERWK D FDXVH DQG HIIHFW RI VWUHVV E\ WKHVH SDUWLFLSDQWV ZKR ZHUH UHFUXLWHG WKURXJK WKH QHZVOHWWHU RI D VXSSRUW QHWZRUN IRU LQGLYLGXDOV ZLWK JHQLWDO KHUSHV 7ZRWKLUGV RI WKH VDPSOH UHSRUWHG H[SHULHQFLQJ PRUH WKDQ ILYH UHFXUUHQFHV LQ \HDU $OWKRXJK WKH VDPSOH PD\ QRW EH UHSUHVHQWDWLYH RI DOO JHQLWDO KHUSHV SDWLHQWV LW GRHV UHIOHFW WKH SHUFHSWLRQV RI D VXEVWDQWLDO QXPEHU RI SDWLHQWV $QRWKHU SRVVLEO\ PRUH UHSUHVHQWDWLYH VDPSOLQJ RI SDWLHQWV b RI ZKRP EHOLHYHG WKH\ ZHUH FXUHG RI JHQLWDO KHUSHV \LHOGHG D VLPLODU ILQGLQJ ,Q WKLV JURXS HPRWLRQDO

PAGE 33

VWUHVV HPHUJHG DV WKH PRVW LPSRUWDQW VHOIUHSRUWHG PHFKDQLVP IRU WULJJHULQJ UHFXUUHQFHV FLWHG E\ b RI WKH VDPSOH %LHUPDQ f 5HWURVSHFWLYH 5HVHDUFK 7D\ORU f VWXGLHG OLIH HYHQWV DQG KHUSHV UHFXUUHQFHV LQ D VDPSOH RI IHPDOH VWXGHQWV RI ZKRP KDG JHQLWDO KHUSHV DQG VHUYHG DV FRQWUROV $OO ZHUH DGPLQLVWHUHG WKH /LIH ([SHULHQFHV 6XUYH\ DQG D *HQLWDO +HUSHV 4XHVWLRQQDLUH :RPHQ ZLWK JHQLWDO KHUSHV GLG QRW UHSRUW D KLJKHU IUHTXHQF\ RI VWUHVVIXO HYHQWV RYHU WKH SULRU \HDU WKDQ ZRPHQ ZLWKRXW KHUSHV :LWKLQ WKH KHUSHV JURXSV KRZHYHU ZRPHQ ZLWK D KLJK UDWH RI UHFXUUHQFH IRXU RU PRUH LQ SUHYLRXV \HDUf ZHUH FRPSDUHG WR WKRVH ZLWK D ORZ UHFXUUHQFH UDWH WKUHH RU OHVVf 7KH KLJK UHFXUUHUV UHSRUWHG KDYLQJ VLJQLILFDQWO\ PRUH QHJDWLYH HYHQWV WKDQ WKH ORZ UHFXUUHUV 'XH WR WKH UHWURVSHFWLYH QDWXUH RI WKH VWXG\ LW LV LPSRVVLEOH WR NQRZ ZKHWKHU KLJK OHYHOV RI QHJDWLYH HYHQWV ZHUH D FDXVH RU HIIHFW RI KHUSHV UHDFWLYDWLRQV ,Q WKH VDPH VWXG\ b RI WKH ZRPHQ LQGLFDWHG WKDW KDYLQJ JHQLWDO KHUSHV FDXVHG HPRWLRQDO VWUHVV $GGLWLRQDOO\ b RI WKH KHUSHV SDWLHQWV UHFDOOHG H[SHULHQFLQJ VWUHVV WKH ZHHN SULRU WR D UHFHQW RFFXUUHQFH ZKLOH RQO\ b RI QRUPDO FRQWUROV UHFDOOHG VWUHVV DW D FRPSDUDEOH SHULRG 7KH SRWHQWLDO UHWURVSHFWLYH ELDV LV DSSDUHQW ,Q DGGLWLRQ WR SDWLHQWVn H[SHFWDWLRQV UHJDUGLQJ VWUHVV DQG LOOQHVV WKH\ KDG D GUDPDWLF HYHQW KHUSHV HSLVRGHf WR WLH WKHLU UHFROOHFWLRQV WR ZKLOH FRQWURO VXEMHFWV PD\ QRW KDYH KDG D WHPSRUDOO\ FRQWLJXRXV HYHQW WR DLG LQ WKHLU UHFDOO RI VWUHVVHV RFFXUULQJ DW WKDW WLPH 7KHVH ILQGLQJV DUH VXSSRUWLYH RI D VWUHVVUHFXUUHQFH UHODWLRQVKLS EXW DUH OLPLWHG E\ UHWURVSHFWLYH ELDV

PAGE 34

WKH ODFN RI REMHFWLYH GRFXPHQWDWLRQ RI WKH LOOQHVV PHDVXUH DQG WKH UHVWULFWHG QDWXUH RI WKH VDPSOH IHPDOH VWXGHQWVf :DWVRQ f H[WHQGHG 7D\ORUnV f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f LV WKH RQO\ LQYHVWLJDWRU WR FRQVLGHU SRWHQWLDO PRGHUDWRUV RI OLIH VWUHVV DQG KHUSHV UHFXUUHQFHV +H LQFOXGHG PHDVXUHV RI SHUFHLYHG ORFXV RI FRQWURO DQG VRFLDO VXSSRUW LQ KLV UHVHDUFK GHVLJQ 8VLQJ WKH 5RWWHU ,QWHUQDO([WHUQDO &RQWURO 6FDOH 5RWWHU f :DWVRQ f GHPRQVWUDWHG DQ LQWHUDFWLRQ EHWZHHQ ORFXV RI FRQWURO DQG VWUHVV LQ ZKLFK VXEMHFWV ZKR SHUFHLYH LQWHUQDO FRQWURO H[SHULHQFHG IHZHU UHFXUUHQFHV XQGHU KLJK VWUHVV FRQGLWLRQV WKDQ WKRVH ZLWK DQ H[WHUQDO ORFXV RI FRQWURO 7KLV LQWHUDFWLRQ HIIHFW VXJJHVWV D PRGHUDWLQJ LQIOXHQFH RI ORFXV RI FRQWURO RQ WKH QHJDWLYH HIIHFWV RI VWUHVV 6RFLDO VXSSRUW DV PHDVXUHG E\ WRWDO KHOSIXOQHVV RQ WKH 6RFLDO 5HODWLRQVKLS 6FDOH GLG QRW DSSHDU WR KDYH D VLPLODU PRGHUDWLQJ HIIHFW RQ VWUHVV +RZHYHU D PRUH GLUHFW HIIHFW ZDV IRXQG 6XEMHFWV UHSRUWLQJ KLJK OHYHOV RI KHOSIXOQHVV KDG VLJQLILFDQWO\ IHZHU UHFXUUHQFHV WKDQ WKRVH UHSRUWLQJ ORZ KHOSIXOQHVV

PAGE 35

:DWVRQ f 7KLV LQYHVWLJDWLRQ HPSKDVL]HV WKH LPSRUWDQFH RI LQYHVWLJDWLQJ SRWHQWLDO PRGHUDWLQJ IDFWRUV ,W LV OLPLWHG E\ UHOLDQFH RQ UHWURVSHFWLYH UHSRUWLQJ DQG WKH QRQUHSUHVHQWDWLYH QDWXUH RI WKH VDPSOH SULPDULO\ KLJK UHFXUUHUV UHFUXLWHG IURP D VHOIKHOS VXSSRUW JURXS 3URVSHFWLYH 5HVHDUFK $ VHULHV RI SURVSHFWLYH LQYHVWLJDWLRQV KDYH EHHQ XQGHUWDNHQ WR H[DPLQH WKH UHODWLRQVKLS RI SV\FKRVRFLDO YDULDEOHV DQG UHFXUUHQFHV RI RUDO +69 )ULHGPDQ .DWFKHU t %ULJKWPDQ .DWFKHU %ULJKWPDQ /XERUVN\ t 6KLS /XERUVN\ 0LQW] %ULJKWPDQ t .DWFKHU f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f :KHQ RQO\ WKH SV\FKRORJLFDO YDULDEOHV ZHUH FRQVLGHUHG WKH XQKDSS\ IDFWRU DFFRXQWHG IRU b RI WKH YDULDQFH 7KH UHODWLRQVKLS EHWZHHQ SUHGLFWRUV ZDV GLIIHUHQW IRU KHUSHV DQG V\VWHPLF LOOQHVV VXJJHVWLQJ WKDW UHFXUUHQFHV GR QRW UHIOHFW DQ LOOQHVV GLVSRVLWLRQ $ VLJQLILFDQW QHJDWLYH UHODWLRQVKLS RI VRFLDO DVVHWV DQG

PAGE 36

KHUSHV HSLVRGHV LPSOLHV D SRVVLEOH PRGHUDWLQJ HIIHFW RI VRFLDO DVVHWV 7KH \HDU UHWURVSHFWLYH OLIH HYHQWV PHDVXUH ZDV RI PLQLPDO YDOXH LQ DOO DQDO\VHV DQG ZDV GLVUHJDUGHG LQ WKH VXEVHTXHQW LQYHVWLJDWLRQV )ULHGPDQ HWDO /XERUVN\ HWDO f :KLOH OLIH HYHQWV ZHUH QRW DQ LPSRUWDQW YDULDEOH LQ WKLV VWXG\ WKLV PD\ EH H[SODLQHG RQ WKH EDVLV RI PHWKRGRORJ\ 7KH QXPEHU RI VXEMHFWV ZKR ZHUH VHURSRVLWLYH IRU KHUSHV ZDV VPDOO $IWHU DWWULWLRQ RQO\ VXEMHFWV FRPSOHWHG D IXOO \HDU RI SDUWLFLSDWLRQ 7KH VWHSZLVH PXOWLSOH UHJUHVVLRQ SURFHGXUHV ZHUH EDVHG RQ HSLVRGHV RI KHUSHV RQO\ RI ZKLFK ZHUH GRFXPHQWHG E\ H[DPLQDWLRQ DQG YLUDO FXOWXUH 7KH OLIH HYHQWV PHDVXUH PD\ KDYH EHHQ LQVHQVLWLYH GXH WR PHPRU\ GHFOLQH 0RQURH Df WKH IDLOXUH WR FRQVLGHU QHJDWLYH HYHQWV VHSDUDWHO\ 'RKUHQZHQG t 'RKUHQZHQG -RKQVRQ t 6DUDVRQ f RU D ODFN RI DVVRFLDWLRQ EHWZHHQ WHPSRUDOO\ UHPRWH HYHQWV DQG UHFXUUHQFHV +RZHYHU WKH LQYHVWLJDWRUV DQHFGRWDOO\ UHSRUW DQ LQFUHDVH RI KHUSHV UHFXUUHQFHV GXULQJ H[DP WLPH IRU VWXGHQWV .DWFKHU HW DO f $ PRQWK GDLO\ LQYHVWLJDWLRQ RI KHUSHV VHURSRVLWLYH VXEMHFWV IDLOHG WR GHPRQVWUDWH DQ\ V\VWHPDWLF UHODWLRQVKLSV EHWZHHQ PRRG DQG KHUSHV UHDFWLYDWLRQV /XERUVN\ HW DO f +RZHYHU DOPRVW DOO VXEMHFWV UHSRUWHG VWUHVVHV SULRU WR HSLVRGHV ZKHQ UHVSRQGLQJ UHWURVSHFWLYHO\ 7KH DXWKRUV VXJJHVW WKDW E\ KDYLQJ PRRGV UHSRUWHG LQ WKH PRUQLQJ WKH HPRWLRQDO LPSDFW RI WKH GD\ ZDV PLVVHG %XW LW PD\ DOVR EH WKH FDVH WKDW PRRGV DUH QRW QHFHVVDULO\ HTXLYDOHQW WR VWUHVV 7KH UHVHDUFKHUV WKHQ IROORZHG D ODUJHU JURXS RI SDUWLFLSDQWV IRU \HDUV )ULHGPDQ HW DO f GXULQJ ZKLFK RI KDG DW OHDVW

PAGE 37

RQH +69 HSLVRGH 8VLQJ KHUSHV LQFLGHQFH DV WKH GHSHQGHQW YDULDEOH LQ D PXOWLSOH UHJUHVVLRQ DQDO\VLV WKH VRFLDO DQG SV\FKRORJLFDO YDULDEOHV ZHUH IRXQG WR EH RI OHDVW SUHGLFWLYH YDOXH b RI H[SODLQHG YDULDQFHf ELRORJLFDO IDFWRUV GLVHDVH KLVWRU\ DQG LOOQHVVHV HVSHFLDOO\ XSSHU UHVSLUDWRU\ LQIHFWLRQf EHLQJ RI JUHDWHVW YDOXH b RI H[SODLQHG YDULDQFHf 8QIRUWXQDWHO\ WKHVH LQYHVWLJDWRUV GLG QRW LQFOXGH D VWUHVV PHDVXUH 7KH\ UHSOLFDWHG WKH HDUOLHU ILQGLQJ WKDW SUHGLFWRUV ZHUH VSHFLILF IRU KHUSHV DQG XQUHODWHG WR SUHGLFWRUV IRU XSSHU UHVSLUDWRU\ LQIHFWLRQ )ULHGPDQ HW DO .DWFKHU HW DO f 'DLO\ VDPSOHV RI YDJLQDO VHFUHWLRQV IURP ILYH ZRPHQ WKUHH RI ZKRP ZHUH SRVLWLYH IRU JHQLWDO KHUSHV ZHUH FROOHFWHG IRU D PRQWK E\ LQYHVWLJDWRUV RI DV\PSWRPDWLF YLUDO VKHGGLQJ $GDP 'UHVVPDQ .DXIPDQ t 0HOQLFN f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f 7KH GDWD IURP D JURXS RI SV\FKLDWULF SDWLHQWV ZLWK SULPDU\ HSLVRGHV RI JHQLWDO KHUSHV ZKR ZHUH IROORZHG SURVSHFWLYHO\ IRU ZHHNV E\ *ROGPHLHU DQG -RKQVRQ f SURYLGH VRPH VXSSRUW IRU D OLQN

PAGE 38

EHWZHHQ SV\FKRVRFLDO IDFWRUV DQG UHFXUUHQFH 6XEMHFWV FRPSOHWHG D LWHP VFUHHQLQJ PHDVXUH RI SV\FKLDWULF V\PSWRPV WKH *HQHUDO +HDOWK 4XHVWLRQQDLUH RQ ZKLFK WKH\ LQGLFDWHG ZKLFK IHHOLQJV DQG EHKDYLRU V\PSWRPVf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t *ROGVWHLQ 5RJHUV 'XEH\ t 5HLFK f 6XPPDU\ 7R VXPPDUL]H GHVSLWH WKH ZLGHVSUHDG SRSXODU QRWLRQ WKDW VWUHVV DQG UHFXUUHQFHV RI KHUSHV DUH OLQNHG FRQVLVWHQW HPSLULFDO VXSSRUW IRU VXFK D UHODWLRQVKLS LV ODFNLQJ $ IHZ DQHFGRWDO UHSRUWV VXJJHVW D FRQQHFWLRQ EHWZHHQ SV\FKRORJLFDO GLVWUHVV DQG UHFXUUHQFHV %ODQN t %URG\ 6FKQHFN f DQG K\SQRWLF VXJJHVWLRQ LV UHSRUWHG WR

PAGE 39

LQIOXHQFH UHFXUUHQFHV -DQLFNL 8OOPDQ f 6WURQJ SODFHER HIIHFWV KDYH EHHQ DSSDUHQW LQ FRQWUROOHG LQYHVWLJDWLRQV +DPLOWRQ .HUQ f 6XUYH\ UHVHDUFK GHPRQVWUDWHV WKDW SDWLHQWV ZLWK UHFXUUHQW KHUSHV EHOLHYH WKHUH LV D FRQQHFWLRQ ZLWK HPRWLRQDO VWUHVV %LHUPDQ +HOS VXUYH\ f ,Q UHWURVSHFWLYH LQYHVWLJDWLRQV RI SDWLHQWV ZLWK JHQLWDO KHUSHV KLJK UHFXUUHQFH UDWHV KDYH EHHQ DVVRFLDWHG ZLWK KLJK PDMRU OLIH VWUHVV 7D\ORU f DQG LQ RQH VWXG\ QHJDWLYH OLIH VWUHVV DSSHDUV WR KDYH H[HUWHG D ORQJ WHUP HIIHFW RQ UHFXUUHQFH UDWH :DWVRQ f 8QIRUWXQDWHO\ DQHFGRWDO UHSRUWV VXUYH\ UHVHDUFK DQG UHWURVSHFWLYH VWXGLHV DUH DOO VXEMHFW WR GLVWRUWLRQ ,Q RQH SURVSHFWLYH LQYHVWLJDWLRQ RI RUDO KHUSHV UHFXUUHQFHV D QHJDWLYH PRRG IDFWRU FRQWULEXWHG VLJQLILFDQWO\ WR UHFXUUHQFH UDWH EXW D OLIH HYHQWV PHDVXUH SURYHG WR EH XQLPSRUWDQW .DWFKHU HW DO f 0RRGV IDLOHG WR KROG XS DV LPSRUWDQW LQ VXEVHTXHQW SURVSHFWLYH VWXGLHV LQ ZKLFK OLIH VWUHVV ZDV XQIRUWXQDWHO\ QRW PHDVXUHG )UHLGPDQ HW DO /XERUVN\ HW DO f $QHFGRWDOO\ UHODWHG VWUHVV HSLVRGHV ZHUH JUDSKLFDOO\ DVVRFLDWHG ZLWK ODERUDWRU\ HYLGHQFH RI YLUDO VKHGGLQJ IRU WKUHH ZRPHQ $GDPV HW DO f $QG ILQDOO\ SDWLHQWV GHPRQVWUDWLQJ D VWURQJ SRWHQWLDO IRU SV\FKLDWULF SUREOHPV RQ D VFUHHQLQJ GHYLFH VXEVHTXHQWO\ UHSRUWHG PRUH UHFXUUHQFHV WKDQ WKRVH IDOOLQJ EHORZ WKH FULWLFDO FXWRII VFRUH VXJJHVWLQJ D FRQQHFWLRQ EHWZHHQ DQ[LHW\ DQG UHFXUUHQFHV *ROGPHLHU t -RKQVRQ f 6R IDU HPSLULFDO VXSSRUW IRU D VWUHVVKHUSHV DVVRFLDWLRQ LV VXJJHVWLYH UDWKHU WKDQ VWURQJ $ ORQJLWXGLQDO SURVSHFWLYH VWXG\ ZLWK UHJXODU

PAGE 40

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nV SHUFHSWLRQ RI HYHQWV LQ HLWKHU QHJDWLYH RU SRVLWLYH ZD\V $ SHUVRQ ZKR FUDYHV H[FLWHPHQW ZRXOG SUREDEO\ YLHZ WKH HYHQW RI D IODW WLUH HQURXWH WR ZRUN GLIIHUHQWO\ WKDQ D FRXQWHUSDUW ZKR WKULYHV RQ URXWLQHV DQG SUHGLFWDELOLW\ $ SHUVRQ ZKR IHHOV FDUHG IRU DQG VXSSRUWHG E\ IDPLO\ DQG IULHQGV ZLOO SUREDEO\ EH OHVV WKUHDWHQHG E\ FULWLFLVP RQ WKH MRE WKDQ D PRUH VROLWDU\ LQGLYLGXDO 7KH SRLQW LV WKDW WKH VDPH HYHQW FRXOG EH SHUFHLYHG DV D VWUHVVRU E\ RQH LQGLYLGXDO EXW QRW DQRWKHU

PAGE 41

6LPSOLILHG PRGHO RI VRPH RI WKH K\SRWKHWLFDO OHYHOV WKURXJK ZKLFK HQYLURQPHQWDO HYHQWV PD\ DIIHFW ELRORJLFDO HYHQWV DQG XOWLPDWHO\ UHVXOW LQ D KHUSHV UHFXUUHQFH

PAGE 43

2QFH DQ HYHQW LV SHUFHLYHG DV D VWUHVVRU LW PD\ LQLWLDWH D YDULHW\ RI SK\VLRORJLFDO UHVSRQVHV DV WKH LQGLYLGXDO VWUXJJOHV WR DGDSW WR WKH QHZ VLWXDWLRQ -XVW DV PRGHUDWRU YDULDEOHV PD\ LQIOXHQFH D SHUVRQn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f 7R H[SODLQ WKLV SKHQRPHQRQ WZR PDMRU K\SRWKHVHV KDYH EHHQ SURSRVHG ,Q RQH K\SRWKHVLV WKH G\QDPLF VWDWH K\SRWKHVLV D VPDOO QXPEHU RI FHOOV FRQVWDQWO\ UHSOLFDWH YLUXV 6SHFLILF LPPXQH GHIHQVH UHVSRQVHV LQYROYLQJ DQWLERG\ O\PSKRF\WHV DQG LQWHUIHURQ FRQWLQXDOO\ ZRUN WR ORFDOL]H DQG SUHYHQW WKH FOLQLFDO PDQLIHVWDWLRQV RI UHFXUUHQW

PAGE 44

LQIHFWLRQ 6WHYHQV f 7KLV WKHRU\ DOVR FDOOHG WKH VNLQ WULJJHU WKHRU\ SURSRVHV WKDW UHFXUUHQFHV UHVXOW ZKHQ ORFDO GHIHQVHV DUH WHPSRUDULO\ VXSSUHVVHG RU ZKHQ VOLJKW SK\VLRORJLFDO FKDQJHV RWKHU WKDQ LPPXQH PHFKDQLVPVf DOWHU WKH EDODQFH LQ IDYRU RI WKH YLUXV +LOO t %O\WK f 7KH VWDWLF VWDWH K\SRWKHVLV RU JDQJOLRQ WULJJHU WKHRU\ SURSRVHV WKDW WKH YLUXV UHPDLQV LQDFWLYH LQ WKH JDQJOLD XQWLO QHUYRXV V\VWHP VWLPXODWLRQ SURYRNHV LW WR PLJUDWH WRZDUG WKH VNLQ VXUIDFH ,Q WKLV WKHRU\ WKH YLUDO JHQRPH UHPDLQV LQ DQ XQSURGXFWLYH TXLHVFHQW VWDWH 6WHYHQV f $W SUHVHQW WKH ZHLJKW RI HYLGHQFH IDYRUV WKH VWDWLF VWDWH K\SRWKHVLV ,QYHVWLJDWRUV WKDW RQFH EHOLHYHG LQ WKH G\QDPLF K\SRWKHVLV +LOO %O\WK f QRZ FRQVLGHU WKH YLUDO JHQRPH LQWDFW EXW LQDFWLYH LQ WKH JDQJOLRQ ,W LV UHSUHVVHG DQG H[SUHVVHV IHZ LI DQ\ YLUDO DQWLJHQV DQG WKHUHIRUH SURYRNHV QR VLJQLILFDQW LPPXQH UHVSRQVH XQOHVV LQIHFWLRQ LV UHDFWLYDWHG %O\WK t +LOO +LOO %O\WK +DUERXU %HUULH t 7XOOR f 7KH SUHFLVH PHFKDQLVP RI GHUHSUHVVLRQ DOORZLQJ YLUDO H[SUHVVLRQ UHPDLQV D P\VWHU\ 7KHUH LV VRPH LQGLFDWLRQ WKDW LPPXQH V\VWHP VWDWXV LQIOXHQFHV DQ LQGLYLGXDOnV YXOQHUDELOLW\ WR UHFXUUHQW KHUSHV 7KLV LV SDUWLFXODUO\ WUXH IRU FHOOPHGLDWHG LPPXQH UHVSRQVHV DV RSSRVHG WR KXPRUDO LPPXQLW\ 2QFH LQIHFWHG ZLWK KHUSHV VLPSOH[ YLUXV LQGLYLGXDOV GHYHORS QHXWUDOL]LQJ DQWLERGLHV KXPRUDO LPPXQLW\f ZLWKLQ WR ZHHNV RI SULPDU\ LQIHFWLRQ 1DKPLDV t 5RL]PDQ Ff 8QIRUWXQDWHO\ WKH DQWLERGLHV DUH QRW HIIHFWLYH LQ SUHYHQWLQJ UHLQIHFWLRQ RU UHFXUUHQW LQIHFWLRQ LQ HLWKHU KHDOWK\ RU LPPXQRFRPSURPLVHG LQGLYLGXDOV 1DKPLDV t 5RL]PDQ D 5DQG 5DQG 5DVPXVVHQ 3ROODUG $UYLQ t

PAGE 45

0HULJDQ f 7KH KHUSHV YLUXVHV EHLQJ FDSDEOH RI FHOO WR FHOO VSUHDG DUH ZHOO DGDSWHG WR HYDGH DQWLERGLHV ,W KDV EHHQ GHPRQVWUDWHG WKDW WKH WLWHUV RI QHXWUDOL]LQJ DQWLERG\ RQO\ RFFDVLRQDOO\ IOXFWXDWH EHIRUH GXULQJ RU DIWHU UHFXUUHQFHV 1DKPLDV t 5RL]PDQ Ef )OXFWXDWLRQV LQ FHOOPHGLDWHG LPPXQH UHVSRQVLYLW\ RQ WKH RWKHU KDQG DSSHDU WR EH UHODWHG WR UHFXUUHQW KHUSHV LQIHFWLRQV (YLGHQFH IRU WKLV UHODWLRQVKLS FRPHV IURP D YDULHW\ RI VRXUFHV $ GHIHFW LQ FHOOXODU LPPXQLW\ LV SURSRVHG WR DFFRXQW IRU WKH VHYHULW\ RI KHUSHV LQIHFWLRQV LQ WKH QHZERUQ DQG WKH LUQPXQRORJLFDOO\ LPSDLUHG KRVW 1DKPLDV t 5RL]PDQ Ff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t +ROPHV f 6LPLODU UHVXOWV ZHUH REWDLQHG LQ DQ LQYHVWLJDWLRQ FRPSDULQJ WKH O\PSKRF\WH UHVSRQVLYLW\ RI SDWLHQWV ZLWK IUHTXHQWO\ UHFXUULQJ RUDO KHUSHV WR D FRQWURO JURXS RI VHURSRVLWLYH EORRG EDQN GRQRUV QRW UHSRUWLQJ IUHTXHQW KHUSHV UHFXUUHQFHV .LUFKQHU 6FKZHQWHFN 1RUWKRII t 6FKRSI f 5HVSRQVLYLW\ RI WKH WZR JURXSV ZDV VLPLODU GXULQJ LQIHFWLRQ EXW GXULQJ WKH GLVHDVHIUHH LQWHUYDO UHVSRQVLYLW\ RI WKH SDWLHQW JURXS ZDV VLJQLILFDQWO\ ORZHU 7KH VHYHULW\ RI LQIHFWLRQ LQ LPPXQRFRPSURPLVHG SDWLHQWV KDV EHHQ UHODWHG

PAGE 46

VSHFLILFDOO\ WR GHFUHDVHG UHVSRQVLYLW\ RI O\PSKRF\WHV DV ZHOO 0H\HUV )ORXUQR\ t 7KRPDV 5DQG HW DO f $QRWKHU SDUDPHWHU RI FHOOXODU LPPXQLW\ WKH VSHFLILF F\WRWR[LFLW\ RI SHULSKHUDO EORRG PRQRQXFOHDU OHXNRF\WHV 3%0/f WR FHOOV LQIHFWHG ZLWK KHUSHV ZDV IRXQG WR IDOO WR ORZ OHYHOV GXULQJ WKH ZHHN SULRU WR UHFXUUHQW +6< LQ D JURXS RI KHDOWK\ VHURSRVLWLYH DGXOWV VKRZLQJ HQKDQFHG UHDFWLYLW\ GXULQJ DFXWH UHFXUUHQW LQIHFWLRQ 5ROD3OH]F]\QVNL t /LHX f 7FHOOV DQG WKH SHUFHQWDJH RI QDWXUDO F\WRWR[LF O\PSKRF\WHV ZHUH DOVR IRXQG WR GHFUHDVH SULRU WR UHFXUUHQW +69 HSLVRGHV DQG WR LQFUHDVH GXULQJ UHFXUUHQFHV )OXFWXDWLRQV LQ LQWHUIHURQ DQRWKHU FRPSRQHQW RI FHOOXODU LPPXQLW\ KDYH OLNHZLVH EHHQ LPSOLFDWHG LQ UHFXUUHQFHV &RUH\ t +ROPHV 5DQG HW DO f DQG DGPLQLVWUDWLRQ RI LQWHUIHURQ KDV VLJQLILFDQWO\ UHGXFHG WKH ULVN RI UHDFWLYDWLRQ RI SDWLHQWV XQGHUJRLQJ VXUJHU\ RQ WKH WULJHPLQDO JDQJOLRQ 3D]LQ HW DO f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

PAGE 47

VWUHVV DQG UHFXUUHQW KHUSHV $OWHUDWLRQV DV D UHVXOW RI VWUHVV KDYH EHHQ GHPRQVWUDWHG LQ SDUDPHWHUV RI FHOOXODU LPPXQLW\ VXFK DV GHFUHDVHV LQ O\PSKRF\WH UHVSRQVLYHQHVV DQG QDWXUDO NLOOHU FHOO DFWLYLW\ %DUWURS /XFNKXUVW /D]DUXV .LORK t 3HQQ\ /RFNH .LHFROW*ODVHU HWDO f &KDQJHV LQ FHO PHGLDWHG LPPXQH IXQFWLRQV DUH LQ WXUQ UHODWHG WR KHUSHV UHFXUUHQFHV $Q LQGLYLGXDOn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t +LOO f )RU H[DPSOH WKH UHFXUUHQFH UDWH WR ZHHNV DIWHU ERQH PDUURZ WUDQVSODQW LV b 0H\HUV HW DO f ,Q WKH ILUVW PRQWKV IROORZLQJ FDUGLDF WUDQVSODQWDWLRQ b RI VHURSRVLWLYH SDWLHQWV H[SHULHQFHG KHUSHV UHFXUUHQFHV RI SURORQJHG GXUDWLRQ DYHUDJLQJ RYHU PRQWKV LQ GXUDWLRQ 5DQG HW DO f )LIW\ SHUFHQW RI WKH VHURSRVLWLYH UHQDO WUDQVSODQW SDLHQWV GHYHORSHG KHUSHV OHVLRQV RI WR ZHHNV GXUDWLRQ ZLWKLQ D PRQWK IROORZXS SHULRG 3DVV HW DO f 2QH RI WKH HDUO\ LQYHVWLJDWLRQV WR LQGLUHFWO\ LPSOLFDWH VWUHVV DV D IDFWRU LQ WKH UHDFWLYDWLRQ RI +69 ZDV GRQH RQ UDEELWV /DLEVRQ DQG

PAGE 48

.LEULFN f H[SHULPHQWDOO\ LQGXFHG UHFXUUHQW RFXODU KHUSHV LQ ODWHQWO\ LQIHFWHG UDEELWV E\ LQWUDPXVFXODU DGPLQLVWUDWLRQ RI HSLQHSKULQH D KRUPRQH ZHOO NQRZQ WR EH UHOHDVHG LQ H[FHVV GXULQJ SHULRGV RI VWUHVV ,QYHVWLJDWRUV GHPRQVWUDWHG VLJQLILFDQW YLUDO LVRODWLRQ DQG LQIHFWLRQ LQ H[SHULPHQWDOO\ LQRFXODWHG DQLPDOV UHODWLYH WR FRQWURO DQLPDOV 2QH FRXOG SRVWXODWH DQ DXWRQRPLF QHUYRXV V\VWHP OLQN EHWZHHQ VWUHVV DQG UHFXUUHQFH EDVHG RQ WKLV HYLGHQFH 5HDFWLYDWLRQ RI KHUSHV LQ DQLPDOV DQG KXPDQV KDV DOVR EHHQ DFFRPSOLVKHG E\ GLUHFW QHXUDO VWLPXODWLRQ /RZ OHYHOV RI HOHFWULF FXUUHQW GHOLYHUHG YLD HOHFWURGHV LPSODQWHG RYHU WKH WULJHPLQDO JDQJOLRQ RI ODWHQWO\ LQIHFWHG UDEELWV FRQVLVWHQWO\ LQGXFH YLUDO VKHGGLQJ DQG KHUSHV UHFXUUHQFHV GHSHQGLQJ RQ WKH LQWHUYDO EHWZHHQ VWLPXOL *UHHQ 5RVERURXJK t 'XQNHO f ,Q KXPDQV WULJHPLQDO VXUJHU\ DSSHDUV WR H[HUW D GLUHFW WULJJHULQJ HIIHFW %HWZHHQ b RI VHURSRVLWLYH LQGLYLGXDOV ZKR XQGHUJR WULJHPLQDO VXUJHU\ H[SHULHQFH UHFXUUHQFH DQGRU YLUDO VKHGGLQJ ZLWKLQ D ZHHN 7KHVH KLJK LQFLGHQFH UHFXUUHQFHV DUH XQUHPDUNDEOH LQ GXUDWLRQ DQG VHYHULW\ DQG DUH UHVWULFWHG WR WKH DUHD VXSSOLHG E\ WKH VWLPXODWHG RU VHFWLRQHG SRUWLRQ RI WKH WULJHPLQDO VHQVRU\ JDQJOLRQ &DUWRQ t .LOERXUQH 3D]LQ HW DO f ([FHVVLYH VWLPXODWLRQ WR SHULSKHUDO WLVVXH ZKLFK UHFHLYHV VHQVRU\ QHUYHV IURP WKH LQIHFWHG JDQJOLRQ LQGXFHV UHDFWLYDWLRQ LQ DQLPDOV +LOO HW DO f 6XFK VWLPXODWLRQ FRXOG EH DQDORJRXV WR UHDFWLYDWLRQ E\ VXQ ZLQG RU SK\VLFDO LQMXU\ LQ QDWXUDOO\ RFFXUULQJ KHUSHV HSLVRGHV 2QH FRXOG H[WHQG WKH FRQFHSW RI UHDFWLYDWLRQ E\ GLUHFW QHXUDO VWLPXODWLRQ WR LQFOXGH H[FHVVLYH QHXUDO ILULQJ DV PLJKW UHVXOW IURP SV\FKRORJLFDO DV ZHOO DV SK\VLFDO VWLPXODWLRQ 7KXV MXVW

PAGE 49

DV K\SQRVLV RU H[SHFWDQF\ SODFHER HIIHFWVf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

PAGE 50

VXUYHLOODQFH ,W LV DOVR SRVVLEOH WKDW ERWK DUH QHFHVVDU\ WR SURYRNH WKH FOLQLFDO PDQLIHVWDWLRQ RI KHUSHV V\PSWRPDWRORJ\ D WULJJHU WR UHDFWLYDWH WKH YLUXV DQG DQ LPPXQRFRPSURPLVHG HQYLURQPHQW LQ ZKLFK LW PD\ IORXULVK $W WKH SUHVHQW WLPH WKH H[DFW PHFKDQLVPV WKDW FRQWURO WKH ODWHQF\ DQG UHDFWLYDWLRQ RI KHUSHV VLPSOH[ YLUXV LQ KXPDQV DUH XQNQRZQ ,PPXQH UHVSRQVH FRXOG SOD\ D YLWDO UROH EXW WKH SK\VLRORJLFDO VWDWH RI WKH QHXURQ FRXOG DOVR EH D NH\ IDFWRU %O\WK t +LOO f %HFDXVH RI LWV FRPSDWLELOLW\ ZLWK H[LVWLQJ HYLGHQFH WKH VXSSRVLWLRQ WKDW ERWK QHXUDO VWLPXODWLRQ DQG LPPXQRVXSSUHVVLRQ DUH QHFHVVDU\ IRU KHUSHV UHDFWLYDWLRQ LV LQWXLWLYHO\ DSSHDOLQJ +RZHYHU H[DPLQDWLRQV RI WKH YDOLGLW\ RI WKLV DQG WKH RWKHU K\SRWKHVHV SUHVHQW VHULRXV GLIILFXOWLHV :KLOH LW LV NQRZQ WKDW VRPH WULJJHUV DUH PRUH HIIHFWLYH WKDQ RWKHUV HJ WULJHPLQDO VXUJHU\ PRUH IUHTXHQWO\ UHVXOWV LQ UHDFWLYDWLRQ RI +69 WKDQ VXQ H[SRVXUHf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

PAGE 51

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n UHVHDUFK DV PRGHUDWRUV RI OLIH VWUHVV ZHUH DVVHVVHG 7KH XVH RI PXOWLSOH UHJUHVVLRQ WHFKQLTXHV HQDEOHG WKH FODULILFDWLRQ RI WKH UHODWLYH LPSRUWDQFH RI VRFLDO VXSSRUW ORFXV RI FRQWURO DQG DURXVDO VHHNLQJ WHQGHQF\ LQ LQIOXHQFLQJ UHFXUUHQFH UDWHV LQ WKH IDFH RI VWUHVV 7KXV WKH SRVVLELOLW\ RI LGHQWLI\LQJ VXEJURXSV RI LQGLYLGXDOV ZKR ZHUH WKRXJKW WR EH SDUWLFXODUO\ YXOQHUDEOH ZDV H[SORUHG ,Q WKH SUHVHQW VWDWH RI WKH DUW RI SV\FKRELRORJLFDO

PAGE 52

UHVHDUFK LQ ZKLFK WKH VLQJOHFDXVH VLQJOHHIIHFW PRGHO LV DFNQRZOHGJHG DV VLPSOLVWLF ZH PXVW GHSHQG XSRQ FRQYHUJLQJ HYLGHQFH IURP WKH GLIIHUHQW OHYHOV RI PDQ\ GLVFLSOLQHV WR REWDLQ D UHDVRQDEOH SLFWXUH RI WKH PDQ\ LQWHUQDO DQG H[WHUQDO HYHQWV DQG G\QDPLF LQWHUDFWLRQV WKDW XOWLPDWHO\ UHVXOW LQ D GLVHDVH RXWFRPH 6FKZDUW] f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

PAGE 53

GHVLJQHG WR H[SORUH WKH UHODWLRQVKLS EHWZHHQ VWUHVVIXO HYHQWV DQG UHFXUUHQFHV XVLQJ ERWK SURVSHFWLYHO\ DQG FRQFXUUHQWO\ FROOHFWHG LQIRUPDWLRQ 7KH ILUVW K\SRWKHVLV XVLQJ D SURVSHFWLYH DSSURDFK FRQFHUQV WKH VWUHQJWK RI WKH UHODWLRQVKLS EHWZHHQ DQ LQGLYLGXDOnV H[SHULHQFH RI PDMRU OLIH HYHQWV DQG VXEVHTXHQW H[SHULHQFH RI KHUSHV UHFXUUHQFHV 6SHFLILFDOO\ WKH ILUVW K\SRWKHVLV VWDWHV 7KH RFFXUUHQFH RI XQGHVLUDEOH OLIH HYHQWV SULRU WR HQWU\ LQWR WKH VWXG\ ZLOO EH SRVLWLYHO\ UHODWHG WR KHUSHV UHFXUUHQFHV GXULQJ WKH PRQWK SHULRG RI LQYHVWLJDWLRQ 7KH QH[W WZR K\SRWKHVHV FRQFHUQ WKH FRQFXUUHQW H[SHULHQFH RI ERWK PDMRU DQG PLQRU OLIH HYHQWV DQG KHUSHV UHFXUUHQFHV 6SHFLILFDOO\ 7KH RFFXUUHQFH RI XQGHVLUDEOH OLIH HYHQWV GXULQJ WKH PRQWK SHULRG RI LQYHVWLJDWLRQ ZLOO EH SRVLWLYHO\ UHODWHG WR KHUSHV UHFXUUHQFHV RYHU WKH VDPH SHULRG RI WLPH 7KH RFFXUUHQFH RI PRQWKO\ KDVVOHV ZLOO EH SRVLWLYHO\ UHODWHG WR KHUSHV UHFXUUHQFHV GXULQJ WKH SHULRG RI LQYHVWLJDWLRQ $GGLWLRQDOO\ LW ZDV K\SRWKHVL]HG WKDW WKUHH PRGHUDWRU YDULDEOHV VRFLDO VXSSRUW ORFXV RI FRQWURO DQG DURXVDO VHHNLQJ WHQGHQF\ ZRXOG LQWHUDFW ZLWK HDFK RWKHU LQ VXFK D ZD\ WKDW DQ LQGLYLGXDOnV YXOQHUDELOLW\ WR KHUSHV UHFXUUHQFHV XQGHU KLJK VWUHVV FRQGLWLRQV ZRXOG EH SUHGLFWDEOH RQ WKH EDVLV RI VFRUHV RQ WKH PRGHUDWRU YDULDEOHV 7KDW LV 7KH IROORZLQJ YDULDEOHV ZKLFK KDYH SUHYLRXVO\ EHHQ VKRZQ WR PRGHUDWH WKH HIIHFWV RI OLIH VWUHVV ZLOO EH UHODWHG WR +69 UHFXUUHQFHV LQ WKH IROORZLQJ ZD\V Df /RZ OHYHOV RI VRFLDO VXSSRUW D SHUFHSWLRQ RI HYHQWV DV H[WHUQDOO\ FRQWUROOHG DQG D WHQGHQF\ WR DYRLG DURXVLQJ

PAGE 54

VLWXDWLRQV ZLOO EH DVVRFLDWHG ZLWK PRUH IUHTXHQW +69 UHFXUUHQFHV Ef 7KH DQWLWKHVLV RI WKH DERYH ZLOO VHUYH WR SURWHFW LQGLYLGXDOV IURP IUHTXHQW +69 UHFXUUHQFHV 7KDW LV LQGLYLGXDOV ZLWK D KLJK OHYHO RI VRFLDO VXSSRUW ZLWK DQ LQWHUQDO SHUFHSWLRQ RI FRQWURO ZKR WHQG WR EH DURXVDO VHHNLQJ ZLOO KDYH IHZHU +69 UHFXUUHQFHV Ff 7KH SUREDEOH IUHTXHQF\ RI UHFXUUHQFH IRU D JLYHQ LQGLYLGXDO ZLOO EH SUHGLFWDEOH RQ WKH EDVLV RI VFRUHV RQ WKH PRGHUDWRU YDULDEOHV

PAGE 55

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nV PRQWKV RI SDUWLFLSDWLRQ WR UHODWH WR GRFXPHQWHG UHFXUUHQFHV IRU WKH VDPH SHULRG RI WLPH 7KLV GHVLJQ SURYLGHG WKH RSSRUWXQLW\ WR DYRLG GLVWRUWLRQ GXH WR UHWURVSHFWLYH ELDV LQ DQ LQYHVWLJDWLRQ RI WKH UHODWLRQVKLS RI PDMRU DQG PLQRU VWUHVVIXO HYHQWV DQG KHUSHV UHFXUUHQFHV 7KH DVVHVVPHQW RI VXEMHFWVn VHQVH RI SHUFHLYHG FRQWURO WHQGHQF\ WRZDUG DURXVDO VHHNLQJ DQG VRFLDO VXSSRUW SURYLGHG DQ RSSRUWXQLW\ WR HYDOXDWH WKH K\SRWKHVL]HG VWUHVVPRGHUDWLQJ HIIHFWV RI WKHVH YDULDEOHV ,Q DGGLWLRQ WR VWXG\LQJ WKH HIIHFWV RI WKH SUHVXPHG PRGHUDWRU YDULDEOHV VHSDUDWHO\ PXOWLYDULDWH VWDWLVWLFDO WHFKQLTXHV PDGH LW SRVVLEOH WR HYDOXDWH WKH PDQQHU LQ ZKLFK WKHVH YDULDEOHV UHODWHG LQ DQ DGGLWLYH RU LQWHUDFWLYH PDQQHU WR SRVVLEO\ HQKDQFH RU

PAGE 56

GLPLQLVK WKH UHVLVWDQFH RI LQGLYLGXDOV WR UHFXUUHQW HSLVRGHV RI JHQLWDO KHUSHV LQ WKH IDFH RI OLIH VWUHVV 6DPSOH 6HOHFWLRQ 6XEMHFWV ZHUH UHFUXLWHG RYHU D PRQWK SHULRG 0DUFK -XO\ f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

PAGE 57

0HDVXUHV 6WUHVV 0HDVXUHV 7KH RFFXUUHQFH RI PDMRU DQG PLQRU HYHQWV LQ WKH OLYHV RI VXEMHFWV ZDV DVVHVVHG E\ VHOIUHSRUW TXHVWLRQQDLUHV 6XEMHFWV FRPSOHWHG WKH TXHVWLRQQDLUHV XSRQ HQUROOPHQW DQG PRQWKO\ GXULQJ WKH VWXG\ /LIH ([SHULHQFHV 6XUYH\ /(6f 7KLV TXHVWLRQQDLUH ZDV XVHG WR HYDOXDWH WKH H[SHULHQFH RI VXEMHFWV ZLWK PDMRU OLIH HYHQWV 7KH /(6 LV FRPSULVHG RI OLIH HYHQWV RI ZKLFK DUH ILOOLQV DQG WKH ODVW RI ZKLFK DUH VSHFLILFDOO\ UHOHYDQW WR D FROOHJH SRSXODWLRQ 6DUDVRQ -RKQVRQ t 6LHJHO f 6XEMHFWV LQGLFDWH WKH IUHTXHQF\ RI RFFXUUHQFH RI HDFK RI WKH HYHQWV RYHU D VSHFLILHG WLPH SHULRG 7KH\ DOVR UDWH WKH LPSDFW RI HDFK HYHQW RQ D VFDOH IURP H[WUHPHO\ QHJDWLYHf WR H[WUHPHO\ SRVLWLYHf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f DQG JHQLWDO KHUSHV LQ SDUWLFXODU 7D\ORU :DWVRQ f 7KLV LV FRQVLVWHQW ZLWK WKH OLWHUDWXUH RQ DGMXVWPHQW 0XHOOHU HW DO 9LQRNXU t 6HO]HU f $OWKRXJK LPSDFW VXEMHFW ZHLJKWHGf VFRUHV ZHUH DYDLODEOH WKH PHDVXUH RI LQWHUHVW LQ WKLV LQYHVWLJDWLRQ ZDV WKH IUHTXHQF\ RI XQGHVLUDEOH

PAGE 58

HYHQWV 6HYHUDO VWXGLHV SURYLGLQJ GDWD RQ ERWK VLPSOH IUHTXHQF\ VFRUHV DQG ZHLJKWHG VFRUHV KDYH VKRZQ WKHP WR EH HTXDOO\ SUHGLFWLYH ZKHUH JURXS GDWD DUH FRQFHUQHG 0RQURH E 6NLQQHU t /HL =LPPHUPDQ f 'HYHORSHUV RI WKH VFDOH KDYH SURYLGHG LQIRUPDWLRQ UHJDUGLQJ VLJQLILFDQW SRVLWLYH FRUUHODWLRQV ZLWK VHYHUDO GLIIHUHQW PHDVXUHV RI DQ[LHW\ QRQFRPIRUPLW\ GLVFRPIRUW GHSUHVVLRQ ORFXV RI FRQWURO DQG D QRQVLJQLILFDQW FRUUHODWLRQ ZLWK WKH 0DUORZH &URZQH 6RFLDO 'HVLUDELOLW\ 6FDOH WR GHPRQVWUDWH YDOLGLW\ $GGLWLRQDOO\ WKH VFDOH KDV EHHQ IRXQG WR GLVFULPLQDWH EHWZHHQ QRUPDO FRQWUROV DQG FRXQVHOLQJ FOLHQWV 6DUDVRQ HW DO f 7KLV VFDOH KDV EHHQ XVHG LQ WZR UHWURVSHFWLYH LQYHVWLJDWLRQV RI VXEMHFWV ZLWK UHFXUUHQW KHUSHV ,Q RQH WKH VFDOH GLVFULPLQDWHG EHWZHHQ VXEMHFWV H[SHULHQFLQJ KLJK DQG ORZ UHFXUUHQFHV 7D\ORU f ,Q WKH RWKHU VLJQLILFDQW SRVLWLYH FRUUHODWLRQV ZHUH IRXQG EHWZHHQ WKH IUHTXHQF\ DQG LPSDFW RI XQGHVLUDEOH OLIH HYHQWV RI WKH SDVW \HDU DQG WKH IUHTXHQF\ DQG GXUDWLRQ RI KHUSHV UHFXUUHQFHV IRU WKH SDVW PRQWKV :DWVRQ f +DVVOHV 6FDOH +6f 7KH +6 IRFXVHV RQ WKH PLQRU HYHQWV RI WKH SDVW PRQWK WKH LUULWDWLQJ IUXVWUDWLQJ DQG GLVWUHVVLQJ GHPDQGV FKDUDFWHUL]LQJ GDLO\ WUDQVDFWLRQV ZLWK WKH HQYLURQPHQW .DQQHU HW DO f 7KH LWHPV ZKLFK LQFOXGH VXFK WKLQJV DV FDUH IRU SHW KDYLQJ WR ZDLW LQFRQVLGHUDWH VPRNHUV WRR PDQ\ PHHWLQJV SUREOHPV ZLWK FKLOGUHQ DQG SUHSDULQJ PHDOV DUH FKHFNHG LI SHUFHLYHG DV D KDVVOH DQG UDWHG IURP WR DFFRUGLQJ WR WKH VHYHULW\ RI WKH KDVVOH $ IUHTXHQF\ VFRUH LV REWDLQHG E\ FRXQWLQJ WKH QXPEHU RI KDVVOHV FKHFNHG $Q LPSDFW VFRUH LV REWDLQHG E\ VXPPLQJ WKH KDVVOH UDWLQJV 1RUPV DUH EDVHG RQ PLGGOH FODVV VXEMHFWV UDQJLQJ LQ DJH

PAGE 59

IURP WR ZKR FRPSOHWHG WKH VFDOH IRU FRQVHFXWLYH PRQWKV 7HVWUHWHVW UHOLDELOLW\ RYHU WKDW SHULRG LV UHSRUWHG DV IRU IUHTXHQF\ VFRUHV IRU LPSDFW VFRUHV 6LJQLILFDQW SRVLWLYH FRUUHODWLRQV KDYH EHHQ IRXQG ZLWK D QHJDWLYH DIIHFW VFDOH D OLIH HYHQWV PHDVXUH D V\PSWRP FKHFN OLVW .DQQHU HW DO f DQG D JHQHUDO KHDOWK PHDVXUH 'H/RQJLV HW DO f 0RGHUDWRU 9DULDEOHV 8SRQ HQUROOPHQW LQ DQG WHUPLQDWLRQ IURP WKH UHVHDUFK PRQWKV ODWHU VXEMHFWV FRPSOHWHG LQYHQWRULHV WR DVVHVV HQGXULQJ FKDUDFWHULVWLFV SURSRVHG WR PRGHUDWH WKH HIIHFWV RI VWUHVV RQ LOOQHVV 7KHVH LQFOXGHG PHDVXUHV RI VRFLDO VXSSRUW RI SHUFHSWLRQ RI LQWHUQDO RU H[WHUQDO FRQWUROODELOLW\ RI HYHQWV DQG RI DURXVDO VHHNLQJ WHQGHQF\ /RFXV RI &RQWURO /&f 7KH /& LV D LWHP IRUFHGFKRLFH LQYHQWRU\ LQFOXGLQJ VL[ ILOOHU LWHPV ZKLFK DVVHVVHV WKH GHJUHH WR ZKLFK LQGLYLGXDOV YLHZ HQYLURQPHQWDO HYHQWV DV EHLQJ XQGHU WKHLU SHUVRQDO FRQWURO 5RWWHU f 6XEMHFWV VFRULQJ ORZ RQ WKH /& LQWHUQDOVf WHQG WR SHUFHLYH HYHQWV DV EHLQJ FRQWUROODEOH E\ WKHLU RZQ DFWLRQV ZKLOH WKRVH VFRULQJ KLJK H[WHUQDOVf WHQG WR YLHZ HYHQWV DV W EHLQJ LQIOXHQFHG E\ IDFWRUV RWKHU WKDQ WKHPVHOYHV ,W LV D PHDVXUH RI JHQHUDOL]HG H[SHFWDQF\ RU EHOLHI UDWKHU WKDQ SUHIHUHQFH 7KH PHDVXUH LV ZLGHO\ XVHG DQG H[WHQVLYH GDWD DUH DYDLODEOH UHJDUGLQJ LQWHUQDO FRQVLVWHQF\ WHVWUHWHVW UHOLDELOLW\ DQG GLVFULPLQDQW YDOLGLW\ &RUUHODWLRQV ZLWK VRFLDO GHVLUDELOLW\ DQG LQWHOOLJHQFH DUH ORZ 0HDQV DQG VWDQGDUG GHYLDWLRQV RQ WKH VFDOH DUH SURYLGHG IRU

PAGE 60

QXPHURXV JURXSV LQFOXGLQJ ODUJH JURXSV RI XQLYHUVLW\ VWXGHQWV 5RWWHU f $ QXPEHU RI VWXGLHV LQYHVWLJDWLQJ SV\FKRORJLFDO DGMXVWPHQW DQG LOOQHVV DV UHODWHG WR OLIH VWUHVV SURYLGH VXSSRUW IRU WKH UROH RI SHUFHLYHG FRQWURO DV D PRGHUDWRU RI OLIH VWUHVV 7KDW LV DQ H[WHUQDO /& VFRUH LV DVVRFLDWHG ZLWK LQVWDELOLW\ DQG LOOQHVV XQGHU FRQGLWLRQV RI LQFUHDVHG OLIH VWUHVV -RKQVRQ t 6DUDVRQ 0DQXFN HW DO 7RYHV HW DO f $URXVDO 6HHNLQJ $6f $URXVDO VHHNLQJ LV GHILQHG DV WKH WHQGHQF\ RI LQGLYLGXDOV WR HQJDJH LQ RU DYRLG VLWXDWLRQV ZKLFK LQFUHDVH VWLPXODWLRQ WR D SHUVRQDO RSWLPDO OHYHO 7KH $6 VFDOH ZDV GHYHORSHG VSHFLILFDOO\ WR DVVHVV LQGLYLGXDO GLIIHUHQFHV LQ WKLV WUDLW 0HKUDELDQ t 5XVVHOO f 6XEMHFWV DUH LQVWUXFWHG WR LQGLFDWH WKHLU GHJUHH RI DJUHHPHQW RU GLVDJUHHPHQW RQ D SRLQW VFDOHf WR HDFK RI LWHPV 0HDQV DQG VWDQGDUG GHYLDWLRQV DUH UHSRUWHG IRU WZR ODUJH VDPSOHV RI FROOHJH VWXGHQWV $GHTXDWH LQWHUQDO FRQVLVWHQF\ DQG UHOLDELOLW\ KDYH EHHQ GHPRQVWUDWHG 7KH $6 VFDOH FRUUHODWHG SRVLWLYHO\ ZLWK D PHDVXUH RI H[WURYHUVLRQ DQG QHJDWLYHO\ ZLWK PHDVXUHV RI DQ[LHW\ &RUUHODWLRQV ZLWK DQ[LHW\ DUH VLJQLILFDQW EXW ZHDN LW LV QRW VLPSO\ DQ DQ[LHW\ PHDVXUHf /RZ FRUUHODWLRQV KDYH EHHQ IRXQG ZLWK VRFLDO GHVLUDELOLW\ DQG VXFFRUDQFH 0HKUDELDQ t 5XVVHOO f ,Q DQ LQYHVWLJDWLRQ RI WKH UHODWLRQVKLS EHWZHHQ OLIH VWUHVV DQG PHDVXUHV RI DQ[LHW\ GHSUHVVLRQ DQG KRVWLOLW\ D VLJQLILFDQW UHODWLRQVKLS ZDV IRXQG RQO\ IRU VXEMHFWV ORZ RQ WKH DURXVDO VHHNLQJ GLPHQVLRQ 1HJDWLYH OLIH FKDQJH LQ VXEMHFWV ZKR ZHUH ORZ LQ DURXVDO VHHNLQJ UHVXOWHG LQ JUHDWHU DQ[LHW\ DQG KRVWLOLW\ 7KLV VXJJHVWV WKDW WKH

PAGE 61

GLPHQVLRQ SOD\V D PRGHUDWLQJ UROH RQ OLIH VWUHVV -RKQVRQ 6DUDVRQ t 6LHJHO f 6RFLDO 6XSSRUW 4XHVWLRQQDLUH 664f 6RFLDO VXSSRUW PD\ EH GHILQHG DV WKH GHJUHH WR ZKLFK LQGLYLGXDOV KDYH DFFHVV WR VRFLDO UHVRXUFHV RU UHODWLRQVKLSV XSRQ ZKLFK WKH\ FDQ UHO\ SDUWLFXODUO\ LQ WLPHV RI QHHG $ SHUVRQnV VDWLVIDFWLRQ ZLWK WKH TXDOLW\ RI VXFK UHVRXUFHV LV DQ LPSRUWDQW GLPHQVLRQ LQ DGGLWLRQ WR WKH QXPEHU RI SHUVRQV LQ D VXSSRUW V\VWHP 7KH 664 PHDVXUHV WKH VL]H DQG VDWLVIDFWLRQ UDWLQJ RI DQ LQGLYLGXDOnV VRFLDO QHWZRUN 6DUDVRQ /HYLQH %DVKDP t 6DUDVRQ f ,W FRQVLVWV RI WZRSDUW LWHPV RQH SDUW WR LQGLFDWH WKH QXPEHU RI VXSSRUW SHRSOH DYDLODEOH XQGHU SDUWLFXODU FLUFXPVWDQFHV WKH RWKHU WR LQGLFDWH WKH UHVSRQGHQWnV VDWLVIDFWLRQ UDWLQJ IRU DYDLODEOH VXSSRUW 1RUPV H[LVW IRU D VDPSOH RI XQLYHUVLW\ VWXGHQWV ,QWHUQDO FRQVLVWHQF\ DQG WHVWUHWHVW UHOLDELOLW\ DUH KLJK &RUUHODWLRQV ZLWK VFDOHV RI PDODGMXVWPHQW VKRZ DQ LQYHUVH UHODWLRQVKLS VXJJHVWLQJ WKDW SHRSOH ZKR KDYH IHZHU DQG OHVV VDWLVI\LQJ VRFLDO VXSSRUWV DUH PRUH OLNHO\ WR EH DQ[LRXV PRUH ODELOH HPRWLRQDOO\ DQG PRUH SHVVLPLVWLF WKDQ RWKHUV 6XEMHFWV ORZ RQ WKH 664 WHQG WR KDYH DQ H[WHUQDO ORFXV RI FRQWURO DQG EH UHODWLYHO\ ORZ LQ VHOIHVWHHP nf 7KHUH KDYH EHHQ VHYHUDO VWXGLHV ZKLFK SURYLGH HYLGHQFH IRU VRFLDO VXSSRUW DV D PRGHUDWRU RI OLIH VWUHVV GH $UDXMR HW DO +RWDOLQJ HW DO 1XFNROOV HW DO 6FKDHIHU HW DO f 7KRVH ORZ LQ VRFLDO VXSSRUW HYLGHQFHG PRUH PDODGMXVWPHQW DQG LOO KHDOWK XQGHU FLUFXPVWDQFHV RI KLJK OLIH VWUHVV 7KHUHIRUH KLJK OHYHOV RI VRFLDO VXSSRUW PD\ SOD\ D VWUHVVEXIIHULQJ UROH ,Q WKH

PAGE 62

SUHVHQW LQYHVWLJDWLRQ WKH 664 ZDV PRGLILHG WR LQFOXGH D SDLU RI TXHVWLRQV UHJDUGLQJ WR ZKRP D VXEMHFW KDG GLVFORVHG KHVKH KDG KHUSHV DQG OHYHO RI VDWLVIDFWLRQ ZLWK VXSSRUW UHFHLYHG 3K\VLRORJLFDO 0HDVXUHV 6FUHHQLQJ SURFHGXUH +69 DQWLERG\ 6LQFH HQUROOPHQW LQ WKH VWXG\ ZDV OLPLWHG WR WKRVH ZKR ZHUH KHUSHV VLPSOH[ YLUXV +69f VHURSRVLWLYH VXEMHFWV ZHUH VFUHHQHG LQ RUGHU WR HQVXUH WKDW WKH\ KDG KDG SULRU H[SRVXUH WR +69 7KLV ZRUN ZDV SHUIRUPHG LQ D YLURORJ\ ODERUDWRU\ DQG ZDV GRQH E\ D PLFURWLWHU FRPSOHPHQW IL[DWLRQ &)f PHWKRG SUHYLRXVO\ GHVFULEHG 5DQG .UDPHU t -RKQVRQ f 1R DWWHPSW ZDV PDGH WR GLVFULPLQDWH EHWZHHQ DQWLERGLHV WR KHUSHV VLPSOH[ YLUXV +69f DQG KHUSHV VLPSOH[ YLUXV +69f EHFDXVH Df VHUD WKDW LV VHURSRVLWLYH E\ &) LV VHURSRVLWLYH IRU ERWK +69 DQG +69 E\ DQ DFFHSWHG GLVFULPLQDWLRQ WHFKQLTXH (/,6$f DQG Ef DOWKRXJK +69 XVXDOO\ FDXVHV JHQLWDO KHUSHV +69 LV DOVR D IUHTXHQW FDXVH 'HSHQGHQW YDULDEOH 9LUXV LVRODWLRQ DQG WUDQVSRUW $OO VXEMHFWV ZHUH LQVWUXFWHG WR FDOO WKH &OLQLFDO 5HVHDUFK &HQWHU &5&f DV VRRQ DV WKH\ ZHUH DZDUH RI D UHFXUUHQW JHQLWDO KHUSHV OHVLRQ $Q DSSRLQWPHQW ZDV PDGH IRU D FXOWXUH WR EH REWDLQHG ZLWKLQ KRXUV E\ WUDLQHG L SHUVRQQHO LQ WKH &5& 7KH &5& SHUVRQQHO ZHUH DYDLODEOH RQ D KRXU GDLO\ EDVLV &XOWXUHV IRU +69 ZHUH REWDLQHG E\ VZDEELQJ WKH EDVH RI XQURRIHG YHVLFXODU RU XOFHUDWHG OHVLRQV ZLWK D FRWWRQ VZDE 6ZDEV ZHUH LPPHUVHG LQ WUDQVSRUW PHGLD >PLQLPDO HVVHQWLDO PHGLXP 0(0f VXSSOHPHQWHG ZLWK PP JOXWDPLQH X SHQLFLOOLQ XJPO

PAGE 63

VWUHSWRP\FLQ XJPO JHQWDPLFLQ XJPO DPSKRWHULFLQ % DQG b JHODWLQ@ DQG KHOG DW r& XQWLO WUDQVSRUWHG WR WKH ODERUDWRU\ SUHYLRXV ZRUN LQ WKH ODERUDWRU\ XQSXEOLVKHGf DQG HOVHZKHUH KDV VKRZQ WKLV PHWKRG GRHV QRW UHVXOW LQ DQ\ VLJQLILFDQW GHFUHDVH LQ UHFRYHU\ RI +69 %HWWROL %UHZHU 2[WRE\ $]LGL t *XLQDQ
PAGE 64

7KH\ ZHUH SURYLGHG ZLWK D 9LURFXOW VZDE DQG WUDLQHG WR REWDLQ D VHOIn FXOWXUH LQ WKH HYHQW WKDW LW ZDV LPSRVVLEOH WR JHW WR WKH &5& ZLWKLQ KRXUV $ FXOWXUH FRXOG EH VDIHO\ NHSW LQ WKH UHIULJHUDWRU IRU WR GD\V XQWLO WUDQVSRUW WR WKH FOLQLF ZDV SRVVLEOH $ VDPSOH FRS\ RI WKH /(6 ZDV SURYLGHG ZLWK D FDOHQGDU RQ ZKLFK WR UHFRUG PDMRU OLIH HYHQWV 7KLV HDVHG WKH ILOOLQJ RXW RI TXHVWLRQQDLUHV HDFK PRQWK DQG KHOSHG WR DYRLG RYHUODS RU GXSOLFDWLRQ RI PDMRU OLIH HYHQWV IURP PRQWK WR PRQWK 6XEMHFWV ZHUH WKHQ JLYHQ D SDFNHW RI LQLWLDO TXHVWLRQQDLUHV DQG LQVWUXFWHG LQ WKH SURFHGXUH IRU FRPSOHWLRQ 7KH SDFNHW FRQWDLQHG WKH /(6 IRU WKH SDVW PRQWK DQG \HDU SHULRGVf WKH +6 /& $6 DQG 664 TXHVWLRQQDLUHV $Q DSSRLQWPHQW ZDV PDGH IRU WKH VHFRQG FRQWDFW LQ WR ZHHNV WLPH ZKHQ WKH EORRG VFUHHQLQJ UHVXOWV ZRXOG EH DYDLODEOH 6HFRQG &RQWDFW 6XEMHFWV UHWXUQHG ZLWK WKH FRPSOHWHG TXHVWLRQQDLUHV RI WKH LQLWLDO SDFNHW 7KHLU TXHVWLRQV ZHUH DQVZHUHG DQG WKH\ ZHUH JLYHQ D VHFRQG RSSRUWXQLW\ WR VLJQ WKH FRQVHQW IRUP WR VLJQLI\ FRQWLQXLQJ FRPPLWPHQW WR WKH LQYHVWLJDWLRQ DW D WLPH ZKHQ EORRG VFUHHQLQJ UHVXOWV ZHUH NQRZQ DQG WKH\ ZHUH WKRURXJKO\ IDPLOLDU ZLWK DOO SURFHGXUHV DQG TXHVWLRQQDLUHV 6XEMHFWV ZHUH QRW FRQVLGHUHG IXOO\ HQUROOHG XQWLO WKHLU VHFRQG VLJQDWXUH ZDV REWDLQHG &RPSOHWHG TXHVWLRQQDLUHV ZHUH UHYLHZHG IRU SURSHU FRPSOHWLRQ DQG D VWUXFWXUHG LQWHUYLHZ ZDV FRQGXFWHG WR REWDLQ GHPRJUDSKLF LQIRUPDWLRQ DQG WKH VXEMHFWVn KLVWRU\ RI KHUSHV 3OHDVH VHH $SSHQGL[ $ IRU D FRS\ RI WKH VWUXFWXUHG LQWHUYLHZf

PAGE 65

&RQWLQXLQJ &RQWDFW &RQWLQXLQJ FRQWDFW ZDV E\ PDLO SKRQH DQG FOLQLF YLVLWV ,Q WKH ODVW ZHHN RI HDFK PRQWK VXEMHFWV UHFHLYHG E\ PDLO D SDFNHW FRQWDLQLQJ SUHFRGHG TXHVWLRQQDLUHV /(6 DQG +6f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

PAGE 66

&+$37(5 7+5(( 5(68/76 6DPSOH &KDUDFWHUV VWLHV 7KH RULJLQDO VDPSOH FRQVLVWHG RI LQGLYLGXDOV )LYH ZHUH H[FOXGHG EHFDXVH WKH\ GLG QRW KDYH +69 DQWLERGLHV LQ WKH LQLWLDO VFUHHQLQJ DQG WKXV ZHUH QRW VXEMHFW WR UHFXUUHQW +69 $WWULWLRQ RFFXUUHG IRU WKH IROORZLQJ UHDVRQV Df PRYHG MA Ef WRR EXV\ QB Ff ORVW LQWHUHVW MU Gf UHFHLYHG H[SHULPHQWDO YDFFLQH QB O Hf EHOLHYHG SDUWLFLSDWLRQ ZRUVHQHG FRQGLWLRQ QB O If YLVXDO KDQGLFDS BQB O DQG Jf ORVW WR IROORZXS QB &RQVLGHULQJ RQO\ WKH VHURSRVLWLYH VXEMHFWV WKLV UHSUHVHQWV D ORVV RI RU b RI WKH HOLJLEOH VXEMHFWV 7KH ILQDO VDPSOH FRQVLVWHG RI VXEMHFWV RI ZKRP ZHUH IHPDOH bf DQG RI ZKRP ZHUH PDOH bf $YHUDJH DJH ZDV 6' f ZLWK D UDQJH RI WR (GXFDWLRQDO OHYHO ZDV KLJK ZLWK WKH DYHUDJH QXPEHU RI \HDUV RI HGXFDWLRQ EHLQJ UDQJH f 6L[W\WKUHH bf RI WKH SDUWLFLSDQWV ZHUH FXUUHQWO\ LQ FROOHJH RU JUDGXDWH VFKRRO 7KH PDMRULW\ RI WKH SDUWLFLSDQWV bf ZHUH VLQJOH DOWKRXJK bf UHSRUWHG KDYLQJ D UHJXODU VH[ SDUWQHU 6XEMHFWV UHSRUWHG KDYLQJ KHUSHV DQ DYHUDJH RI \HDUV PRQWKV UDQJH RI PRQWK WR \HDUVf )RU WKH \HDU SULRU WR WKH VWXG\ b UHSRUWHG IRXU RU PRUH UHFXUUHQFHV b UHSRUWHG RQH WR WKUHH UHFXUUHQFHV DQG b UHSRUWHG QRQH

PAGE 67

8QGHVLUDEOH /LIH (YHQWV 7KH PHDVXUHV RI PDMRU DQG PLQRU OLIH HYHQWV LQ WKLV VWXG\ HDFK SURGXFHG WZR VFRUHV RI SRWHQWLDO UHOHYDQFH WR WKH SUHVHQW LQYHVWLJDWLRQ D VLPSOH IUHTXHQF\ RI QHJDWLYH HYHQWV VFRUH DQG DQ LPSDFW VFRUH UHIOHFWLQJ VXEMHFWVn LQGLYLGXDO ZHLJKWLQJ RI WKHVH HYHQWV /LIH HYHQWV VFRUHV RI LPSRUWDQFH LQFOXGHG D PRQWK UHFDOO RI PDMRU HYHQWV RFFXUULQJ MXVW SULRU WR HQUROOPHQW DQG PRQWKO\ UHSRUWV RI OLIH HYHQWV ZKLFK ZHUH VXPPHG WR SURGXFH D PRQWK FRQFXUUHQW WRWDO RI HYHQWV H[SHULHQFHG GXULQJ WKH LQYHVWLJDWLRQ 7KH +6 ZDV DOVR FRPSOHWHG PRQWKO\ DQG VXPPHG WR SURYLGH D PRQWK WRWDO KDVVOHV LQGH[ 6HH $SSHQGL[ & IRU VXEMHFWVn VFRUHV RQ DOO YDULDEOHVf &RQVLVWHQW ZLWK SUHYLRXV LQYHVWLJDWLRQV =LPPHUPDQ f 3HDUVRQ SURGXFWPRPHQW FRUUHODWLRQV RI IUHTXHQF\ DQG LPSDFW VFRUHV UHVXOWHG LQ H[WUHPHO\ KLJK FRUUHODWLRQV 7KH FRUUHODWLRQ EHWZHHQ WKH PRQWK UHWURVSHFWLYH /(6 IUHTXHQF\ DQG LPSDFW VFRUHV ZDV e e f EHWZHHQ /(6 FRQFXUUHQW IUHTXHQF\ DQG LPSDFW e e f DQG EHWZHHQ FRQFXUUHQW +6 IUHTXHQF\ DQG LPSDFW e e f $OWKRXJK LW PLJKW EH DUJXHG WKDW WKH ZHLJKWHG PHDVXUHV PD\ EH PRUH VHQVLWLYH WR DQ LQGLYLGXDOnV SHUFHSWLRQ RI WKH VWUHVVIXO QHVV RI PDMRU DQG PLQRU HYHQWV JLYHQ WKH UHGXQGDQF\ LQ LQIRUPDWLRQ SURYLGHG E\ WKHVH LQGLFHV WKH\ ZHUH QRW LQFOXGHG LQ IXUWKHU JURXS DQDO\VHV 6LPSOH IUHTXHQF\ FRXQWV RI QHJDWLYH HYHQWV ZHUH XVHG IRU ERWK WKH /(6 DQG +6 PHDVXUHV DV WKH\ ZHUH VHHQ DV SURYLGLQJ WKH PRVW REMHFWLYH DQG VWUDLJKWIRUZDUG LQGLFHV RI PDMRU DQG PLQRU OLIH FKDQJHV UHVSHFWLYHO\

PAGE 68

7DEOH SURYLGHV PHDQV DQG VWDQGDUG GHYLDWLRQV REWDLQHG RQ VWUHVV PHDVXUHV IRU WKH PRQWKV SULRU WR WKH VWXG\ SDVWf DQG IRU WKH PRQWKV GXULQJ WKH VWXG\ FRQFXUUHQWf 7DEOH 0HDQV DQG 6WDQGDUG 'HYLDWLRQV RI 3DVW DQG &RQFXUUHQW 6WUHVV 0HDVXUHV 7LPH 3HULRG 6WUHVV 0HDVXUH 3DVW PRQWKV QB &RQFXUUHQW PRQWKV QB )UHTXHQF\ RI 0DMRU 1HJDWLYH /LIH (YHQWV f f ,PSDFW RI 0DMRU 1HJDWLYH /LIH (YHQWV f f +DVVOHV )UHTXHQF\ f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

PAGE 69

SHULRGV GLIIHUHG 6HFRQGO\ WKH GLVFUHSDQF\ PD\ EH DWWULEXWHG WR PHPRU\ ORVV IRU HYHQWV WKDW RFFXUUHG RYHU WKH PRUH UHPRWH PRQWK SHULRG SULRU WR HQWU\ LQWR WKH VWXG\ 0RQURH D 3D\NHO f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f EXW ZRXOG RQO\ KDYH EHHQ FRXQWHG RQFH RQ WKH PRQWK UHFDOO RI PDMRU HYHQWV H[SHULHQFHG SULRU WR WKH VWXG\ ,W LV RI LQWHUHVW WR FRPSDUH /(6 VFRUHV RI WKH SUHVHQW VDPSOH ZLWK WKRVH RI RWKHU JURXSV 1RUPDWLYH GDWD DUH SURYLGHG RQ WKH /(6 LPSDFW UDWLQJV RI PDOH DQG IHPDOH VWXGHQWV HQUROOHG LQ LQWURGXFWRU\ SV\FKRORJ\ FODVVHV 6DUDVRQ HWDO f $OWKRXJK LPSDFW VFRUHV ZHUH QRW XVHG LQ DQDO\VHV LQ WKH SUHVHQW VWXG\ WKH\ ZHUH DYDLODEOH DQG DUH XVHG KHUH IRU FRPSDULVRQ SXUSRVHV VLQFH

PAGE 70

QRUPDWLYH GDWD KDYH EHHQ SURYLGHG IRU /(6 LPSDFW EXW QRW /(6 IUHTXHQF\ VFRUHV 7KH PHDQ PRQWK UHFDOO QHJDWLYH LPSDFW VFRUH RI KHUSHV VXEMHFWV ZDV A'B f DV FRQWUDVWHG WR D PHDQ VFRUH RI 6' f IRU PDOHV RU 6'B f IRU IHPDOHV IRU D PRQWK UHFDOO DGPLQLVWUDWLRQ RI WKH /(6 2QH ZRXOG KDYH H[SHFWHG WKH ILJXUHV WR EH UHYHUVHG VLQFH WKH GDWD RQ WKH QRUPDWLYH VDPSOH DUH EDVHG RQ D WLPH SHULRG WKDW LV WZLFH DV ORQJ DV RSSRVHG WR PRQWKVf 7KH GLIIHUHQFH LV HYHQ PRUH VWULNLQJ IRU FRQFXUUHQW LQIRUPDWLRQ ZLWK DQ DYHUDJH IUHTXHQF\ RI PDMRU QHJDWLYH OLIH HYHQWV RI 6' f 7KH ODWWHU DV DOUHDG\ GLVFXVVHG LQFOXGHV LQIRUPDWLRQ RQ IUHTXHQWO\ RFFXUULQJ DQG SRWHQWLDOO\ UHSHDWHG PDMRU HYHQWV $XWKRUV RI WKH RULJLQDO PHDVXUH VXJJHVW WKDW WKH /(6 VFRUH RI WKH VWXGHQW QRUPDWLYHf SRSXODWLRQ PD\ EH ORZ UHODWLYH WR VXEMHFWV IURP WKH JHQHUDO SRSXODWLRQ 6DUDVRQ HW DO f ,W DSSHDUV WKDW WKH SUHVHQW PRUH KHWHURJHQHRXV VDPSOH KDV SHUFHLYHG UHODWLYHO\ PRUH VWUHVV WKDQ WKH QRUPDWLYH JURXS 6FRUHV RQ WKH +6 DUH PRUH GLUHFWO\ FRPSDUDEOH WR QRUPDWLYH GDWD DV ERWK DUH EDVHG RQ PRQWKO\ DGPLQLVWUDWLRQV .DQQHU HW DO f 7KH QRUPDWLYH VXEMHFWV ZHUH ROGHU UDQJLQJ LQ DJH IURP WR DV RSSRVHG WR WKH DJH UDQJH LQ WKH SUHVHQW VDPSOH RI WR 7KH DYHUDJH PRQWKO\ KDVVOHV IUHTXHQF\ VFRUH RI WKH QRUPDWLYH JURXS ZDV ZKLFK LV YHU\ VLPLODU WR WKH DYHUDJH IRU WKH KHUSHV VDPSOH RI

PAGE 71

3UHVXPHG 0RGHUDWRU 9DULDEOHV ,QIRUPDWLRQ UHJDUGLQJ WKH KHUSHV VDPSOH DV D ZKROH RQ DURXVDO VHHNLQJ WHQGHQF\ ORFXV RI FRQWURO DQG VRFLDO VXSSRUW LV SURYLGHG KHUH IRU FRPSDULVRQ SXUSRVHV 6XEMHFWVn VFRUHV RQ DURXVDO VHHNLQJ WHQGHQF\ ZHUH DSSUR[LPDWHO\ QRUPDOO\ GLVWULEXWHG ZLWK D PHDQ RI QB f 7KLV LV VOLJKWO\ KLJKHU WKDQ WKH DYHUDJH RI UHSRUWHG IRU WKH QRUPDWLYH VDPSOH RI VXEMHFWV 0HKUDELDQ t 5XVVHOO f 7KH VWDQGDUG GHYLDWLRQV RI WKH WZR VDPSOHV DUH VLPLODU IRU WKH KHUSHV VXEMHFWV DQG IRU WKH QRUPDWLYH JURXS /RFXV RI &RQWURO /&f VFRUHV ZHUH DOVR DSSUR[LPDWHO\ QRUPDOO\ GLVWULEXWHG ZLWK D PHDQ VFRUH RI AO QB f $YHUDJH VFRUHV LQ WKH OLWHUDWXUH UDQJH IURP 6'B QB f IRU 3HDFH &RUSV YROXQWHHUV WR 6'B QA f IRU FROOHJH VWXGHQWV 5RWWHU f $OWKRXJK WKH VWDQGDUG GHYLDWLRQV DUH VLPLODU WKH DYHUDJH VFRUH RI WKH KHUSHV VDPSOH LV VOLJKWO\ KLJKHU LQGLFDWLQJ WKDW DV D JURXS WKH SUHVHQW VXEMHFWV WHQG WR EH PRUH H[WHUQDOO\ RULHQWHG WKDQ RWKHUV :KHQ WKH DYHUDJH 6RFLDO 6XSSRUW 6DWLVIDFWLRQ 666f VFRUH LV FRPSDUHG WR QRUPDWLYH GDWD LW LQLWLDOO\ DSSHDUV WKDW WKH KHUSHV VXEMHFWVn 666 VFRUHV PD\ EH VOLJKWO\ ORZHU 7KH DYHUDJH 666 VFRUH IRU KHUSHV VXEMHFW LV 6' QB f IRU WKH QRUPDWLYH JURXS RI XQGHUJUDGXDWHV WKH PHDQ ZDV 6'BQRW SURYLGHG 6DUDVRQ HW DO f +RZHYHU WKH GLVWULEXWLRQ RI VFRUHV IURP WKH KHUSHV VXEMHFWV UDQJHG IURP WR DQG ZDV KLJKO\ VNHZHG ZLWK PRVW RI WKH VFRUHV JURXSHG WRJHWKHU DW WKH YHU\ VDWLVILHG HQG RI WKH VFDOH

PAGE 72

PHGLDQ f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e f 7KH +6 VFRUHV ZHUH UHWDLQHG IRU DQDO\VHV RI K\SRWKHVHV VSHFLILFDOO\ FRQFHUQLQJ PLQRU HYHQWV 'HILQLWLRQ RI 5HFXUUHQFH 7UXH KHUSHV UHFXUUHQW OHVLRQV GR QRW DOZD\V SURGXFH SRVLWLYH FXOWXUH UHVXOWV LQ WKH ODERUDWRU\ GXH WR D UHODWLYHO\ VKRUW GXUDWLRQ RI YLUDO VKHGGLQJ &RUH\ DQG +ROPHV f UHSRUW REWDLQLQJ D b FXOWXUH SRVLWLYH UDWH IURP UHFXUUHQW OHVLRQV ,I WKH GHILQLWLRQ RI WKH GHSHQGHQW PHDVXUH ZDV UHVWULFWHG WR FXOWXUHSRVLWLYH OHVLRQV DV PDQ\ DV b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

PAGE 73

GHILQLWLRQV GHPRQVWUDWHG D KLJK GHJUHH RI DVVRFLDWLRQ ZLWK HDFK RWKHU 6SHDUPDQ FRUUHODWLRQ FRHIILFLHQWV RI WKH QXPEHU RI UHFRUGHG UHFXUUHQFHV DQG WKH QXPEHU RI SRVLWLYH FXOWXUHV ZLWK WKH QXPEHU RI UHFXUUHQFHV JLYHQ RQH SRVLWLYH ZHUH DQG ERWK VWDWLVWLFDOO\ VLJQLILFDQW DW WKH OHYHO )XUWKHUPRUH FRUUHODWLRQV RI HDFK RI WKH UHFXUUHQFH PHDVXUHV ZLWK WKH VWUHVV PHDVXUHV ZHUH YHU\ VLPLODU WR HDFK RWKHU 5HIHU WR $SSHQGL[ IRU FRUUHODWLRQ FRHIILFLHQWV RI WKH VWUHVV DQG UHFXUUHQFH PHDVXUHVf 7KHUHIRUH VXEVHTXHQW DQDO\VHV LQFOXGH DOO UHSRUWHG UHFXUUHQFHV IRU VXEMHFWV IRU ZKRP DW OHDVW RQH SRVLWLYH ZDV GRFXPHQWHG +HUHDIWHU XQOHVV RWKHUZLVH VWDWHG WKH WHUP UHFXUUHQFHV ZLOO UHIHU WR WKRVH PHHWLQJ WKH UHTXLUHPHQWV RI WKLV GHILQLWLRQ 'XULQJ WKH VWXG\ RI WKH WRWDO VXEMHFWV UHSRUWHG KHUSHV OHVLRQV 2I WKHVH ZHUH FXOWXUHG bf DQG RI WKRVH FXOWXUHG ZHUH SRVLWLYH bf 7KLV SRVLWLYH UDWH FRPSDUHV IDYRUDEO\ WR WKDW UHSRUWHG HDUOLHU &RUH\ t +ROPHV f 7KHUH ZHUH VXEMHFWV ZKRVH OHVLRQV UHPDLQHG XQFRQILUPHG E\ ODERUDWRU\ GLDJQRVLV 8VLQJ UHFXUUHQFHV DV GHILQHG DERYH WKH K\SRWKHVHV RI WKH SUHVHQW LQYHVWLJDWLRQ ZHUH WHVWHG )RU HDVH RI SUHVHQWDWLRQ UHVXOWV FRQFHUQLQJ WKH H[LVWHQFH DQG H[WHQW RI D GLUHFW UHODWLRQVKLS EHWZHHQ VWUHVV DQG UHFXUUHQFHV ZLOO EH LQWURGXFHG ILUVW /LQHDU FRUUHODWLRQ FRHIILFLHQWV ZHUH XVHG WR H[DPLQH WKLV DVVRFLDWLRQ )ROORZLQJ WKDW DQDO\VHV SHUWDLQLQJ WR WKH FRQGLWLRQV XQGHU ZKLFK VXFK D UHODWLRQVKLS H[LVWV DUH SUHVHQWHG ,QFOXGHG KHUH DUH WKH UHVXOWV RI PXOWLYDULDWH DQDO\VHV GHVLJQHG WR LQYHVWLJDWH WKH FRPELQHG LQIOXHQFH RI WKH VWUHVV PHDVXUHV DQG SUHVXPHG PRGHUDWRU YDULDEOHV )LQDOO\ ILQGLQJV RI

PAGE 74

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e e f %HFDXVH WKLV UHVXOW ZDV VXUSULVLQJ LQ OLJKW RI WZR SUHYLRXV UHWURVSHFWLYH LQYHVWLJDWLRQV 7D\ORU :DWVRQ f WKH DQDO\VLV ZDV UHSHDWHG IRU WKH VXEVHW RI VXEMHFWV ZKR H[SHULHQFHG UHFXUUHQFHV GXULQJ WKH VWXG\ HOLPLQDWLQJ DOO VXEMHFWV ZLWK ]HUR UHFXUUHQFHV 7KLV SURGXFHG D VXEVDPSOH RI UHFXUULQJ VXEMHFWV PRUH VLPLODU WR WKH SUHYLRXV VDPSOHV VWXGLHG 7KH UHVXOW ZDV DOVR QRQVLJQLILFDQW e e f 7KHVH UHVXOWV LQGLFDWH D WRWDO ODFN RI DVVRFLDWLRQ EHWZHHQ UHFHQW XQGHVLUDEOH OLIH HYHQWV DQG VXEVHTXHQW KHUSHV UHFXUUHQFHV SURYLGLQJ QR VXSSRUW IRU +\SRWKHVLV ,Q FRQWUDVW D VLJQLILFDQW SRVLWLYH UHODWLRQVKLS ZDV REWDLQHG EHWZHHQ PRQWKO\ WRWDOV RI WKH IUHTXHQF\ RI UHSRUWHG XQGHVLUDEOH OLIH

PAGE 75

HYHQWV VXPPHG RYHU WKH PRQWK SHULRG RI WKH LQYHVWLJDWLRQ DQG UHFXUUHQFHV H[SHULHQFHG RYHU WKH VDPH SHULRG RI WLPH 7KH FRUUHODWLRQ REWDLQHG ZDV e e f 7KLV VLJQLILFDQW SRVLWLYH FRUUHODWLRQ LQGLFDWHV WKDW LQFUHDVLQJ OHYHOV RI UHSRUWHG QHJDWLYH OLIH HYHQWV ZHUH DVVRFLDWHG ZLWK LQFUHDVLQJ UHFXUUHQFH UDWHV ZKHQ FRQFXUUHQWO\ PRQLWRUHG 7KXV VRPH VXSSRUW ZDV REWDLQHG IRU WKH VHFRQG K\SRWKHVLV WKDW VWUHVV DQG UHFXUUHQFHV DUH FRQFXUUHQWO\ UHODWHG :LWK UHJDUG WR +\SRWKHVLV FRQFHUQLQJ PLQRU OLIH HYHQWV H[SHULHQFHG RYHU WKH VDPH SHULRG RI WLPH QR VLJQLILFDQW UHODWLRQVKLS ZDV GHPRQVWUDWHG 7KH FRUUHODWLRQ RI WKH WRWDO +6 VFRUHV UHSRUWHG PRQWKO\ DQG UHFXUUHQFHV ZDV RQO\ e f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

PAGE 76

PXOWLYDULDWH UHJUHVVLRQ DQDO\VHV PDNH EHWWHU XVH RI WKH FRQWLQXRXV QDWXUH RI WKH GDWD DQG FDQ SURYLGH D PRUH SUHFLVH GHVFULSWLRQ RI WKH UHODWLRQVKLSV DPRQJ WKH YDULDEOHV /LQQ f ,Q OLQHDU PRGHOV PXOWLYDULDWH WHFKQLTXHV DOORZ IRU WKH DVVHVVPHQW RI WKH HIIHFW RI HDFK YDULDEOH ZKLOH VWDWLVWLFDOO\ KROGLQJ RWKHU YDULDEOHV RI LQWHUHVW FRQVWDQW 7KH\ DOVR DOORZ IRU QRQOLQHDU PRGHOV ZKLFK FDQ LQFOXGH LQWHUDFWLRQ WHUPV DQG WKHUHIRUH PDNH LW SRVVLEOH WR DVVHVV WKH PRGHUDWLQJ HIIHFWV RI RQH YDULDEOH RQ DQRWKHU 6$6 8VHUnV *XLGH f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f QRQUHFXUUHUV ZKR H[SHULHQFHG QR GRFXPHQWHG UHFXUUHQFHV GXULQJ WKH VWXG\ Ef UHFXUUHUV ZKR KDG RQH RU WZR UHFXUUHQFHV DQG Ff UHFXUUHUV ZKR KDG WKUHH RU PRUH UHFXUUHQFHV /RJLVWLF UHJUHVVLRQ DQDO\VLV ZDV FKRVHQ DV WKH PRVW DSSURSULDWH DQDO\VLV IRU XVH ZLWK WKLV RUGHUHG FDWHJRULFDO GDWD +DUUHOO f $ ORJLVWLF UHJUHVVLRQ FDQ EH XVHG WR HVWLPDWH QRQOLQHDU UHODWLRQVKLSV

PAGE 77

DQG WDNHV LQWR DFFRXQW WKH RUGHUHG QDWXUH RI WKH UHFXUUHQFH FDWHJRULHV $ILIL t &ODUN +DUUHOO f $ IRUZDUG VWHSZLVH ORJLVWLF UHJUHVVLRQ SURFHGXUH ZDV SHUIRUPHG WR LGHQWLI\ ZKLFK LI DQ\ YDULDEOHV DQGRU LQWHUDFWLRQV ZHUH UHODWHG WR UHFXUUHQFH JURXSV 6LQJOH LQGHSHQGHQW YDULDEOHV DYDLODEOH WR HQWHU WKH PRGHO FRQVLVWHG RI WKH WRWDO QXPEHU RI QHJDWLYH PDMRU OLIH HYHQWV UHSRUWHG PRQWKO\ DQG VXPPHG RYHU WKH PRQWKV RI WKH VWXG\ /(617f WKH WRWDO QXPEHU RI KDVVOHV UHSRUWHG PRQWKO\ DQG VXPPHG RYHU WKH PRQWKV RI WKH VWXG\ +67f DQG WKH IROORZLQJ WKUHH YDULDEOHV ZKLFK ZHUH PHDVXUHG DW WKH EHJLQQLQJ RI WKH VWXG\ ORFXV RI FRQWURO /&f VDWLVIDFWLRQ ZLWK VRFLDO VXSSRUW 666f DQG DURXVDO VHHNLQJ WHQGHQF\ $6f 7ZRIDFWRU LQWHUDFWLRQV ZKLFK LQFOXGHG DW OHDVW RQH VWUHVV PHDVXUH +67 [ /& +67 [ 666 +67 [ $6 /(617 [ /& /(617 [ 666 /(617 [ $6 /(617 [ +67f DQG WKUHHIDFWRU LQWHUDFWLRQV ZKLFK LQFOXGHG RQH VWUHVV PHDVXUH +67 [ /& [ 666 +67 [ /& [ $6 +67 [ $6 [ 666 /(617 [ /& [ 666 /(617 [ /& [ $6 /(617 [ $6 [ 666f ZHUH DOO DYDLODEOH WR HQWHU WKH PRGHO LI LPSRUWDQW LQ SUHGLFWLQJ UHFXUUHQFHV 7KH OHYHO RI VLJQLILFDQFH UHTXLUHG IRU D YDULDEOH WR HQWHU DQG UHPDLQ LQ WKH PRGHO ZDV 7KH RQO\ WHUP VHOHFWHG DV VLJQLILFDQW LQ WKLV PDQQHU ZDV WKH /(617 [ 666 LQWHUDFWLRQ M.26f 7KH ORJLVWLF UHJUHVVLRQ SURFHGXUH ZDV UHSHDWHG ZLWK WZR YDULDWLRQV Df 6XEMHFWV ZHUH HOLPLQDWHG ZKRVH FXOWXUHV ZHUH QHYHU GRFXPHQWHG E\ SRVLWLYH FXOWXUH Q f ,Q PRVW FDVHV WKHVH FXOWXUHV ZHUH GRFXPHQWHG LQ WKH VHQVH WKDW D &5& QXUVH KDG MXGJHG WKH OHVLRQ ZRUWK\ RI FXOWXUH 7KHUHIRUH VRPH ZHUH WUXH UHFXUUHQFHV IRU ZKLFK DQ DWWHPSW WR LVRODWH WKH YLUXV IDLOHG Ef 6H[ ZDV LQFOXGHG DV D YDULDEOH LQ RUGHU WR VWDWLVWLFDOO\ FRQWURO IRU D VLJQLILFDQW VH[ GLIIHUHQFH RQ UHVSRQVHV WR WKH 664 BW f e f 5HPRYLQJ WKH SUREDEOH UHFXUUHUV IURP WKH JURXS RI QRQUHFXUUHUV DQG FRQWUROOLQJ IRU VH[ GLIIHUHQFHV GLG QRW DOWHU WKH ORJLVWLF UHJUHVVLRQ VROXWLRQ

PAGE 78

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b RI WKH VXEMHFWVn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

PAGE 79

UHPDLQLQJ RQO\ LI WKH\ FRQWULEXWH LPSRUWDQWO\ LQ UHODWLRQ WR DOO RWKHU YDULDEOHV LQ WKH PRGHO 6$6 8VHUnV *XLGH f 7KH VDPH LQGHSHQGHQW YDULDEOHV ZHUH DYDLODEOH WR HQWHU WKH PRGHO DV LQ WKH LQLWLDO ORJLVWLF UHJUHVVLRQ 7KH IRUZDUG VHOHFWLRQ WHFKQLTXH ZLWK D SB OHYHO RI UHTXLUHG IRU YDULDEOHV WR HQWHU DQG UHPDLQ LQ WKH HTXDWLRQ NHSW RQO\ RQH WHUP WKH /(617 [ /& LQWHUDFWLRQ 7KH DPRXQW RI YDULDQFH DFFRXQWHG IRU E\ WKLV LQWHUDFWLRQ ZDV RQO\ b $ IROORZXS UHJUHVVLRQ DQDO\VLV ZKLFK LQFOXGHG RQO\ /& /(617 DQG WKHLU LQWHUDFWLRQ ZDV SHUIRUPHG 7KLV UHVXOWHG LQ D SUHGLFWLRQ HTXDWLRQ ZKLFK LQGLFDWHG WKDW IRU LQGLYLGXDOV ZLWK D PRUH H[WHUQDO ORFXV RI FRQWURO KLJK YDOXHV RI /&f JUHDWHU OLIH VWUHVV UHVXOWHG LQ PRUH UHFXUUHQFHV 7KH RSSRVLWH UHODWLRQVKLS KHOG WUXH IRU PRUH LQWHUQDOO\ RULHQWHG LQGLYLGXDOV 7KH SRLQW DW ZKLFK WKH VORSH RQ /(617 FKDQJHG ZDV DSSUR[LPDWHO\ 6R IRU YDOXHV RI /& JUHDWHU WKDQ /(617 ZDV SRVLWLYHO\ UHODWHG WR UHFXUUHQFHV IRU YDOXHV RI /& OHVV WKDQ /(617 ZDV QHJDWLYHO\ UHODWHG WR UHFXUUHQFHV ,Q WKLV SDUWLFXODU VDPSOH /& VFRUHV UDQJHG IURP WR ZLWK D PHDQ VFRUH RI 6'B f DQG WKH GLVWULEXWLRQ RI /& VFRUHV ZDV QHDU QRUPDO $V SUHYLRXVO\ GLVFXVVHG WKH KHUSHV VXEMHFWV WHQG WR EH PRUH H[WHUQDOO\ RULHQWHG WKDQ QRUPDWLYH JURXSV 5RWWHU f )ROORZXS UHJUHVVLRQ DQDO\VHV ZHUH SHUIRUPHG VHSDUDWHO\ IRU WKUHH JURXSV RI LQGLYLGXDOV EDVHG XSRQ WKHLU /& VFRUHV Df LQGLYLGXDOV ZLWK /& VFRUHV EHWZHHQ DQG /&f Ef WKRVH ZLWK /& VFRUHV EHWZHHQ DQG /&f DQG Ff WKRVH ZLWK /& VFRUHV EHWZHHQ DQG /&f 0HDQ /(617 VFRUHV IRU WKHVH JURXSV ZHUH DQG UHVSHFWLYHO\ 7KH UHODWLRQVKLS EHWZHHQ PDMRU OLIH VWUHVV DQG

PAGE 80

UHFXUUHQFHV ZDV GLIIHUHQW IRU WKHVH JURXSV DV LV LOOXVWUDWHG LQ )LJXUH $V FDQ EH VHHQ E\ H[DPLQDWLRQ RI WKLV ILJXUH WKH UHJUHVVLRQ OLQHV IRU SUHGLFWLQJ WKH QXPEHU RI UHFXUUHQFHV IURP DQ LQGLYLGXDOnV /(617 VFRUH DUH TXLWH GLIIHUHQW LQ VORSH IRU LQGLYLGXDOV ZLWK WKH ORZHVW /& VFRUHV /RZ /& VFRUHV DUH DVVRFLDWHG ZLWK EHOLHI LQ SHUVRQDO LQWHUQDOf FRQWURO RI HYHQWV ZKLOH KLJK VFRUHV DUH DVVRFLDWHG ZLWK EHOLHI LQ H[WHUQDO FRQWURO ,QGLYLGXDOV ZLWK WKH ORZHVW VFRUHV /&f GHPRQVWUDWH D VOLJKWO\ QHJDWLYH QRQVLJQLILFDQW VORSH ZKLOH KLJKHU /& VFRUHV /& DQG /&f GHPRQVWUDWH D SRVLWLYH VORSH ,Q WKH /& DQG /& JURXSV WKH HVWLPDWHG UHODWLRQVKLS ZDV DSSUR[LPDWHO\ WKH VDPH IRU ERWK JURXSV EXW RQO\ WKH VORSH IRU WKH /& JURXS ZDV VLJQLILFDQW e f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

PAGE 81

3UHGLFWHG 1XPEHU RI 5HFXUUHQFHV 1XPEHU RI 0DMRU 1HJDWLYH /LIH (YHQWV 0RQWKO\ 7RWDOV 6XPPHG 2YHU 6L[ 0RQWK 6WXG\ )LJXUH 3UHGLFWHG UHFXUUHQFHV DV D IXQFWLRQ RI PDMRU QHJDWLYH OLIH HYHQWV IRU WKUHH UDQJHV RI /RFXV RI &RQWURO /&f VFRUHV

PAGE 82

7KH 1DWXUH RI WKH /LIH 6WUHVV5HFXUUHQFH 5HODWLRQVKLS 7KH PHWKRG RI GDWD FROOHFWLRQ LQ WKLV LQYHVWLJDWLRQ DOORZHG IRU D PRUH UHILQHG ZLWKLQVXEMHFW DQDO\VLV RI WKH VWUHVVUHFXUUHQFH UHODWLRQVKLS ,QIRUPDWLRQ UHJDUGLQJ OLIH HYHQWV ZDV REWDLQHG DW UHJXODU PRQWKO\ LQWHUYDOV ZLWK UHFXUUHQFHV EHLQJ UHFRUGHG E\ WKH GDWH RI RQVHW 7KLV LQIRUPDWLRQ ZDV XWLOL]HG LQ DQ DWWHPSW WR FODULI\ WKH GLUHFWLRQDOLW\ RI WKH VWUHVVUHFXUUHQFH UHODWLRQVKLS 6SHFLILFDOO\ WKH K\SRWKHVLV ZDV WKDW VWUHVVIXO HYHQWV ZRXOG SUHFHGH UHFXUUHQFHV DQG WKLV ZRXOG EH UHIOHFWHG E\ DQ LQFUHDVHG QXPEHU RI VWUHVVIXO HYHQWV LQ WKH PRQWK SULRU WR WKDW RI D UHFXUUHQFH ,I RQ WKH RWKHU KDQG UHFXUUHQFHV UHVXOWHG LQ LQFUHDVHG VWUHVV WKLV ZRXOG EH UHIOHFWHG LQ DQ LQFUHDVHG QXPEHU RI VWUHVVIXO HYHQWV LQ WKH PRQWK IROORZLQJ WKH UHFXUUHQFH 7R LQYHVWLJDWH WKLV K\SRWKHVLV VWUHVV GLIIHUHQFH VFRUHV ZHUH FRPSXWHG DQG VXPPHG IRU VXEMHFWV ZKR UHSRUWHG UHFXUUHQFHV 7KH GLIIHUHQFH VFRUHV ZHUH GHULYHG E\ VXEWUDFWLQJ WKH VWUHVV VFRUH IRU WKH PRQWK IROORZLQJ WKH PRQWK RI D UHFXUUHQFH IURP WKH VWUHVV VFRUH UHSRUWHG WKH PRQWK SULRU WR D UHFXUUHQFH ,I WKH PHDQ RI WKHVH LQGLYLGXDO GLIIHUHQFH VFRUHV ZDV VLJQLILFDQWO\ GLIIHUHQW IURP ]HUR LQ D SRVLWLYH GLUHFWLRQ LW ZRXOG LQGLFDWH WKDW VWUHVV LV PRUH OLNHO\ WR SUHFHGH WKDQ IROORZ D UHFXUUHQFH DQG VWUHQJWKHQ WKH QDWXUH RI WKH FDXVDO LQIHUHQFH WR EH PDGH $ QRQVLJQLILFDQW RU QHJDWLYH UHVXOW ZRXOG ZHDNHQ WKH FDVH WKDW VWUHVV FDXVHV UHFXUUHQFHV SHUKDSV VXJJHVWLQJ WKDW UHFXUUHQW HSLVRGHV RI KHUSHV UHVXOW LQ LQFUHDVHG VWUHVV RU WKDW VRPH WKLUG IDFWRU LV UHODWHG WR VWUHVV DQG UHFXUUHQFHV FDXVLQJ VLPLODU FKDQJHV WR ERWK

PAGE 83

6LQFH LW LV SRVVLEOH WKDW VWUHVV VFRUHV GXULQJ PRQWKV RI D UHFXUUHQFH FRXOG EH LQIODWHG GXH WR IDFWRUV UHODWHG WR WKH UHFXUUHQFH WKHVH PRQWKV ZHUH DYRLGHG LQ WKH DQDO\VLV 7KDW LV VFRUHV XVHG ZHUH IURP WKH ILUVW FOHDU PRQWK PRQWK ZLWKRXW D UHFXUUHQFHf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 f %HFDXVH WKLV ZDV D ZLWKLQVXEMHFWV DQDO\VLV DQG ZHLJKWHG VFRUHV DUH PRUH VHQVLWLYH WKDQ VLPSOH IUHTXHQF\ VFRUHV WR DQ LQGLYLGXDOVn SHUFHSWLRQ RI WKH VWUHVVIXOQHVV RI HYHQWV WKLV DQDO\VLV ZDV UHSHDWHG XVLQJ LPSDFW VFRUHV $ VLPLODU QRQVLJQLILFDQW

PAGE 84

GLIIHUHQFH VFRUH ZDV REWDLQHG 7KH DYHUDJH LPSDFW GLIIHUHQFH EHWZHHQ WKH PRQWK SUHFHGLQJ DQG WKH PRQWK IROORZLQJ D UHFXUUHQFH ZDV BWB e f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f 7ZHQW\RQH VXEMHFWV UHSRUWHG PRUH VWUHVV SUHFHGLQJ D UHFXUUHQFH UHSRUWHG PRUH VWUHVV IROORZLQJ 7KH UHVW RI WKH VXEMHFWV KDG HTXDO SUHUHFXUUHQFH DQG SRVWUHFXUUHQFH VFRUHV 0F1HPDUnV WHVW ZLWK D FRUUHFWLRQ IRU FRQWLQXLW\ UHVXOWHG LQ D &KLVTXDUH RI ZKLFK LV QRW VLJQLILFDQW DW WKH OHYHO 7KXV QHLWKHU RI WKH ZLWKLQVXEMHFW DQDO\VHV SURYLGHG VXSSRUW IRU WKH QRWLRQ WKDW QHJDWLYH PDMRU OLIH VWUHVV LV D VLJQLILFDQW IDFWRU LQ SUHFLSLWDWLQJ KHUSHV UHFXUUHQFHV

PAGE 85

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

PAGE 86

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t 6WUHXQLQJ f 0RQWKO\ UHSRUWLQJ RI VWUHVVRUV ZDV GHVLJQHG WR DYRLG GLVWRUWLRQ GXH WR IRUJHWWLQJ )XQFK t 0DUVKDOO 0RQURH D 3D\NHO f (SLVRGHV RI KHUSHV REMHFWLYHO\ GRFXPHQWHG E\ YLUDO FXOWXUH FDQQRW EH DWWULEXWHG WR D QHHG WR VHHN DWWHQWLRQ LQ WKH IDFH RI VWUHVV 0HFKDQLF 0LQWHU t .LPEDOO f RU D ORZHUHG WROHUDQFH IRU GLVFRPIRUW GXH WR VWUHVV

PAGE 87

+DUQH\ t %ULJKDP f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t %URG\ -DQLFNL WOOOPDQ f VXUYH\ UHVHDUFK %LHUPDQ +HOS PHPEHUVKLS +69 VXUYH\ f DQG UHWURVSHFWLYH LQYHVWLJDWLRQV 7D\ORU :DWVRQ f 3URVSHFWLYH LQYHVWLJDWLRQV KDYH HLWKHU QRW IRXQG D VLJQLILFDQW DVVRFLDWLRQ SRVVLEO\ GXH WR LQDGHTXDWH PHWKRGRORJ\ .DWFKHU HW DO f RU IDLOHG WR V\VWHPDWLFDOO\ PRQLWRU VWUHVV )UHLGPDQ HW DO /XERUVN\ HW DO f 7KH EHVW HPSLULFDO HYLGHQFH IRU D VWUHVVUHFXUUHQFH DVVRFLDWLRQ KDV EHHQ SURYLGHG E\ 7D\ORU f DQG :DWVRQ f LQ UHWURVSHFWLYH VWXGLHV EDVHG RQ VHOIUHSRUWHG XQGRFXPHQWHG HSLVRGHV RI UHFXUUHQW KHUSHV 7D\ORU f IRXQG WKDW ZRPHQ ZKR UHSRUWHG D KLJK UDWH RI UHFXUUHQFH DOVR UHSRUWHG D KLJKHU IUHTXHQF\ RI VWUHVVIXO HYHQWV RYHU WKH SUHYLRXV \HDU :DWVRQnV f UHSOLFDWLRQ RI WKH ILQGLQJ LQ D ODUJHU VDPSOH RI PDOHV DQG IHPDOHV VWUHQJWKHQV FRQILGHQFH LQ LW 7KH FRUUHODWLRQ EHWZHHQ PDMRU QHJDWLYH OLIH HYHQWV DQG UHFXUUHQFHV IRXQG LQ WKH SUHVHQW VWXG\ UB f LV LQ DJUHHPHQW ZLWK VLJQLILFDQW DQG SRVLWLYHf EXW ZHDNHU WKDQ WKRVH

PAGE 88

IRXQG E\ :DWVRQ U U f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f :KHWKHU UHPRWH RU UHFHQW HYHQWV DIIHFW D SDUWLFXODU LOOQHVV LV DOVR GHSHQGHQW XSRQ WKH FKDUDFWHULVWLHV RI WKH LOOQHVV 7KH WLPLQJ RI HYHQWUHSRUWLQJ LQ OLIHVWUHVV UHVHDUFK KDV EHHQ JHQHUDOO\ D PDWWHU RI FRQYHQLHQFH RU FRQYHQWLRQ ,Q DQ HDUO\ LQYHVWLJDWLRQ E\ 5DKH DQG +ROPHV FLWHG LQ +ROPHV t 0DVXGD f SK\VLFLDQV UHWURVSHFWLYHO\ UHSRUWHG VWUHVVIXO HYHQWV DQG KHDOWK FKDQJHV RYHU D \HDU SHULRG 0RUH UHFHQWO\ LQ OLJKW RI LQIRUPDWLRQ UHJDUGLQJ IRUJHWWLQJ DQG GLVWRUWLRQ )XQFK t

PAGE 89

0DUVKDOO 0RQURH Df VKRUWHU LQWHUYDOV KDYH EHHQ XVHG $ GHFLVLRQ UHJDUGLQJ WKH RSWLPDO UHSRUWLQJ LQWHUYDO WR XVH LQ RUGHU WR REVHUYH D UHODWLRQVKLS FRXOG DOVR EH PDGH FRQVLGHULQJ SK\VLRORJLFDO LQIRUPDWLRQ RI WKH SDUWLFXODU GLVHDVH LQ TXHVWLRQ DQG UHODWHG VWUHVVLOOQHVV UHVHDUFK ILQGLQJV :LWK UHVSHFW WR JHQLWDO KHUSHV WKHUH DUH UHDVRQV WR ORRN ERWK DW ORQJ DQG YHU\ VKRUW ODWHQF\ SHULRGV $ ORQJ ODWHQF\ SHULRG LV VXJJHVWHG E\ :DWVRQnV f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e e f DQG e e f UHVSHFWLYHO\ LQGLFDWLQJ QR VLJQLILFDQW UHODWLRQVKLS EHWZHHQ VWUHVV UHFDOOHG IRU WKH PRQWKV SULRU WR WKH VWXG\ DQG UHFXUUHQFHV LQ WKH ILUVW RU PRQWKV RI LQYHVWLJDWLRQ ,W LV SRVVLEOH WKDW WKH HDUOLHU ILQGLQJ PD\ KDYH EHHQ LQIOXHQFHG E\ UHWURVSHFWLYH ELDV RU WKH KRPRJHQHLW\ RI WKH VDPSOH :DWVRQ f DQG UHVXOWHG LQ D VSXULRXV ILQGLQJ 0RUH UHVHDUFK LV QHHGHG WR UHVROYH WKLV LVVXH $Q DUJXPHQW FDQ EH PDGH RQ WKH EDVLV RI WKH SUHVHQW LQYHVWLJDWLRQ IRU D VKRUWHU ODWHQF\ 7KH VLJQLILFDQW ILQGLQJV ZHUH LQ WKH

PAGE 90

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f DQG 8OOPDQ f EHWZHHQ HPRWLRQDO XSVHW DQG DSSHDUDQFH RI EOLVWHUV 'LUHFW QHXUDO VWLPXODWLRQ VXFK DV RFFXUV

PAGE 91

GXULQJ WULJHPLQDO VXUJHU\ UHOLDEO\ UHVXOWV LQ UHFXUUHQW KHUSHV HSLVRGHV LQ GD\V LQ KXPDQV &DUWRQ t .LOERXUQH 3D]LQ HW DO f ,Q DQ DQLPDO PRGHO HOHFWULFDO VWLPXODWLRQ RI WKH QHUYH SURGXFHV KHUSHV EOLVWHUV ZLWKLQ GD\V *UHHQ HW DO f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f DQG SUHVXPHG VWUHVV PRGHUDWRUV DV ZHOO DV PDMRU OLIH HYHQWV 7KHUH LV D QRWLRQ WKDW WKH LOO HIIHFWV RI PDMRU OLIH VWUHVV PD\ EH PHGLDWHG E\ DQ LQFUHDVH LQ WKH QXPEHU RI KDVVOHV H[SHULHQFHG .DQQHU HW DO f 7KDW LV WKH LOO HIIHFWV RI PDMRU HYHQWV HJ PRYLQJf PD\ EH GXH WR LQFUHDVHG KDVVOHV HJ WURXEOH JHWWLQJ PHDOV EHLQJ ORQHO\ WRR PDQ\ WKLQJV WR GRf ,Q WKH SUHVHQW UHVHDUFK KDVVOHV GLG QRW GHPRQVWUDWH D GLUHFW UHODWLRQVKLS WR UHFXUUHQFHV DQG WKH LQFOXVLRQ RI KDVVOHV GLG QRW HQKDQFH WKH SUHGLFWDELOLW\ RI UHFXUUHQFHV LQ HLWKHU UHJUHVVLRQ DQDO\VLV ,Q WKLV VDPSOH DW OHDVW WKH DVVRFLDWLRQ RI PDMRU QHJDWLYH OLIH HYHQWV DQG UHFXUUHQFHV ZDV LQGHSHQGHQW RI DQ DVVRFLDWLRQ ZLWK KDVVOHV

PAGE 92

:LWK UHJDUG WR WKH PRGHUDWRU YDULDEOHV LW ZDV SUHGLFWHG WKDW WKH\ ZRXOG LQWHUDFW ZLWK HDFK RWKHU LQ VXFK D ZD\ WKDW D SHUVRQn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t 6DUDVRQ f WUHDWPHQW VHHNLQJ 0DQXFN HW DO f DQG UHWURVSHFWLYHO\ UHSRUWHG KHUSHV UHFXUUHQFHV :DWVRQ f ,Q JHQHUDO D UHODWLRQVKLS LV VHHQ EHWZHHQ VWUHVV DQG LOOQHVV RQO\ IRU H[WHUQDOO\ RULHQWHG LQGLYLGXDOV 7KH SUHVHQW VWXG\ IRXQG VXSSRUW IRU WKLV LQ WKH VXEVDPSOH RI VXEMHFWV ZKR KDG UHFXUUHQFHV 1R UHODWLRQVKLS ZDV IRXQG EHWZHHQ VWUHVV DQG UHFXUUHQFHV IRU LQGLYLGXDOV RQ WKH H[WUHPH ORZ HQG RI WKH VFDOH LQWHUQDOVf $ VLJQLILFDQW SRVLWLYH UHODWLRQVKLS ZDV IRXQG EHWZHHQ VWUHVV DQG UHFXUUHQFHV IRU H[WHUQDOV SURYLGLQJ IXUWKHU VXSSRUW IRU ORFXV RI FRQWURO DV D 6WUHVV PRGHUDWRU 6RFLDO VXSSRUW KDV EHHQ ZLGHO\ UHVHDUFKHG E\ LQYHVWLJDWRUV XVLQJ D ZLGH YDULHW\ RI PHDVXUHV ,Q JHQHUDO HLWKHU DQ RYHUDOO EHQHILFLDO HIIHFW RU D EXIIHULQJ UROH LQ UHODWLRQ WR OLIH VWUHVV KDV EHHQ DWWULEXWHG WR VRFLDO VXSSRUW &RKHQ t :LOOV f 7KHVH EHQHILFLDO HIIHFWV KDYH EHHQ PRUH FRQVLVWHQWO\ GHPRQVWUDWHG ZLWK SV\FKRORJLFDO RXWFRPHV WKDQ ZLWK LOOQHVV PHDVXUHV KRZHYHU :DWVRQ f IRXQG D

PAGE 93

GLUHFW HIIHFW RQ JHQLWDO KHUSHV UHFXUUHQFHV VXFK WKDW WKRVH KLJK RQ VRFLDO VXSSRUW SHUFHLYHG KHOSIXOQHVVf KDG VLJQLILFDQWO\ IHZHU UHFXUUHQFHV ,W ZDV LPSRUWDQW WR UHSOLFDWH WKHVH ILQGLQJV LQ D SURVSHFWLYH LQYHVWLJDWLRQ EHFDXVH UHWURVSHFWLYH ELDV DQG UHFUXLWPHQW SURFHGXUHV LQ WKH HDUOLHU VWXG\ OLPLW WKH JHQHUDOL]DELOLW\ RI WKH ILQGLQJV )RUW\ RI WKH VXEMHFWV ZHUH UHFUXLWHG WKURXJK D KHUSHV VHOIKHOS JURXS ZKLFK PD\ KDYH ELDVHG WKH ILQGLQJV UHJDUGLQJ SHUFHLYHG KHOSIXOQHVV :DWVRQ f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n VFRUHV VXJJHVWHG WKDW VXEMHFWV ZHUH HLWKHU YHU\ VDWLVILHG RU YHU\ YHU\ VDWLVILHG ZLWK VXSSRUW UHFHLYHG ,W LV SRVVLEOH WKDW LQGLYLGXDOV LQ WKLV SDUWLFXODU VDPSOH ZHUH SUHGRPLQDQWO\ YHU\ DFWLYH VRFLDOO\ DQG KDSS\ ZLWK WKDW +RZHYHU VXEMHFWV PD\ KDYH RYHUH[WHQGHG WKHPVHOYHV VRFLDOO\ WKXV QHJDWLYHO\ DIIHFWLQJ WKHLU UHVLVWDQFH WR UHFXUUHQFHV 8QIRUWXQDWHO\ WKH UHVWULFWHG QDWXUH RI WKH VDPSOH ZLWK UHJDUG WR VRFLDO VXSSRUW OLPLWV WKH JHQHUDOL]DELOLW\ RI WKHVH ILQGLQJV

PAGE 94

,W ZDV FRQVLGHUHG WKDW VXEMHFWV UHSRUWLQJ KLJK VDWLVIDFWLRQ ZLWK VRFLDO VXSSRUW cQD\ KDYH EHHQ PRUH VH[XDOO\ DFWLYH ZKLFK LQ WXUQ OHG WR PRUH IUHTXHQW UHFXUUHQFHV RI JHQLWDO KHUSHV %HFDXVH LQIRUPDWLRQ ZDV DYDLODEOH RQ WKH IUHTXHQF\ RI VH[XDO DFWLYLW\ RQ D PRQWKO\ EDVLV WKLV SRVVLELOLW\ ZDV HPSLULFDOO\ H[DPLQHG )UHTXHQF\ RI VH[XDO DFWLYLW\ ZDV IRXQG WR EH UHODWHG WR VRFLDO VXSSRUW VDWLVIDFWLRQ e e f EXW WKH QXPEHU RI UHFXUUHQFHV ZDV QHJDWLYHO\ UHODWHG WR IUHTXHQF\ RI VH[XDO DFWLYLW\ e e f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f SURYLGH D XVHIXO W\SRORJ\ RI IRXU VXSSRUW UHVRXUFHV HVWHHP VXSSRUW LQIRUPDWLRQDO VXSSRUW VRFLDO FRPSDQLRQVKLS DQG LQVWUXPHQWDO VXSSRUW 7KH\ VXJJHVW WKDW WKHUH PXVW EH D UHDVRQDEOH PDWFK EHWZHHQ FRSLQJ UHTXLUHPHQWV DQG DYDLODEOH VXSSRUW LQ RUGHU IRU VWUHVV EXIIHULQJ WR RFFXU 5HODWLQJ WKLV WR WKH SUHVHQW ILQGLQJV LW LV FRQFHLYDEOH WKDW UHVSRQGHQWVn UHSRUWHG SHUFHLYHG VDWLVIDFWLRQ RQ WKH 664 LV PRUH UHIOHFWLYH RI WKH W\SH RI VXSSRUW ODEHOHG VRFLDO

PAGE 95

FRPSDQLRQVKLS E\ &RKHQ DQG :LOOV f 7KXV GHVSLWH WKH WHQGHQF\ WR EH TXLWH SOHDVHG ZLWK DYDLODEOH VXSSRUW LW PD\ QRW EH WKH W\SH RI VXSSRUW WKDW DVVLVWV RQH LQ GHDOLQJ ZLWK LQFUHDVHG VWUHVV $V VXJJHVWHG E\ *UDYHV DQG *UDYHV f ZKRVH PHDVXUHV RI LOOQHVV V\PSWRPV DOVR SURYLGHG HYLGHQFH FRQWUDU\ WR D VWUHVVEXIIHULQJ UROH IRU VRFLDO VXSSRUW SHUKDSV WKH REOLJDWRU\ QDWXUH RI GHQVH VRFLDO QHWZRUNV LV PDODGDSWLYH XQGHU FRQGLWLRQV RI OLIH FKDQJH 7RJHWKHU ZLWK WKH ZLGH UDQJH RI ILQGLQJV UHJDUGLQJ VRFLDO VXSSRUW LQ WKH OLWHUDWXUH VHH UHYLHZV E\ &RKHQ t :LOOV 7KRLWV f WKH SUHVHQW UHVXOWV IXUWKHU HPSKDVL]H WKH FRPSOH[LW\ RI WKH SKHQRPHQRQ RI VRFLDO VXSSRUW DQG WKH QHHG IRU PRUH VSHFLILF LQIRUPDWLRQ FRQFHUQLQJ VRXUFH DQG W\SH RI VXSSRUW $URXVDO VHHNLQJ WHQGHQF\ ZDV K\SRWKHVL]HG WR H[HUW D VWUHVV PRGHUDWLQJ UROH RQ WKH EDVLV RI VWXGLHV GHDOLQJ ZLWK DGMXVWPHQW RXWFRPHV 6XEMHFWV ORZ RQ DURXVDO VHHNLQJ WHQGHQF\ KDYH EHHQ VKRZQ WR EH PRUH VXEMHFW WR SV\FKRORJLFDO GLVFRPIRUW XQGHU KLJK VWUHVV FRQGLWLRQV WKDQ KLJK DURXVDO VHHNHUV -RKQVRQ HW DO 6PLWK HW DO f 7KH SUHVHQW LQYHVWLJDWLRQ XQVXFFHVVIXOO\ DWWHPSWHG WR UHSOLFDWH DQG H[WHQG WKRVH ILQGLQJV ZLWK UHJDUG WR D SK\VLFDO RXWFRPH JHQLWDO KHUSHV UHFXUUHQFHVf $OWKRXJK WKH ORJLVWLF UHJUHVVLRQ DQG VWHSZLVH UHJUHVVLRQ SURFHGXUHV UHVXOWHG LQ LQWHUHVWLQJ ILQGLQJV RQO\ b RI WKH YDULDQFH LQ UHFXUUHQFHV ZDV DFFRXQWHG IRU )XUWKHUPRUH EHFDXVH D GLIIHUHQW PRGHUDWRU YDULDEOH ZDV FKRVHQ LQ HDFK UHJUHVVLRQ DQDO\VLV GHSHQGLQJ XSRQ ZKHWKHU WKH IXOO VDPSOH RU RQO\ UHFXUUHUV ZHUH XWLOL]HG LW VXJJHVWV WKH HIIHFWV PD\ QRW EH UREXVW 6RFLDO VXSSRUW ZDV LPSRUWDQW

PAGE 96

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nV DELOLW\ WR FRSH ZLWK VWUHVV DQG WKXV PLWLJDWH WKH LOOQHVVSURGXFLQJ HIIHFWV RI VWUHVVIXO OLIH HYHQWV FRSLQJ VNLOOV VXFK DV SUREOHPVROYLQJ DELOLW\ FRXOG EH PRUH GLUHFWO\ HYDOXDWHG c ,Q DGGLWLRQ WR SV\FKRORJLFDO IDFWRUV WKH UROH RI ELRORJLFDO IDFWRUV VKRXOG EH H[DPLQHG 5HVXOWV RI DQ HDUOLHU SURVSHFWLYH VWXG\ RI RUDO KHUSHV VXJJHVW WKDW ELRORJLFDO IDFWRUV DUH EHWWHU SUHGLFWRUV RI RUDO KHUSHV UHFXUUHQFH WKDQ SV\FKRORJLFDO IDFWRUV LQFOXGLQJ PRRG DQG VRFLDO DVVHWV 'LVHDVH KLVWRU\ DQG LOOQHVVHV DFFRXQWHG IRU RI H[SODLQHG LOOQHVV )ULHGPDQ HW DO f $ PHDVXUH RI LOOQHVV

PAGE 97

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b RI WKH YDULDQFH ZLWK UHJDUG WR JHQLWDO KHUSHV UHFXUUHQFHV UHPDLQV XQH[SODLQHG ,W EHKRRYHV UHVHDUFKHUV DQG FOLQLFLDQV WR FROODERUDWH ZLWK SURIHVVLRQDOV LQ DQ LQWHUGLVFLSOLQDU\ PDQQHU WR IXUWKHU XQWDQJOH WKH FRPSOH[LWLHV RI UHFXUUHQFHV

PAGE 98

$33(1',; $ 6758&785(' ,17(59,(: ,QLWLDO 6WUXFWXUHG ,QWHUYLHZ 'HPRJUDSKLF ,QIRUPDWLRQ 6H[ f ) f 0 $JH 0DULWDO 6WDWXV f 6LQJOH f /LYLQJ ZLWK VLJQLILFDQW RWKHU f 0DUULHG f 6HSDUDWHG f 'LYRUFHG f :LGRZHG +RZ PDQ\ FKLOGUHQ GR \RX KDYH" f f f f f f RU PRUH :KDW OHYHO RI HGXFDWLRQ KDYH \RX FRPSOHWHG" f *UDGH VFKRRO f +LJK VFKRRO JUDGXDWH f 6RPH FROOHJH f &ROOHJH JUDGXDWH f 6RPH JUDGXDWH ZRUN f 0DVWHUnV GHJUHH f 3K' 0' RU RWKHU DGYDQFHG GHJUHH $UH \RX SUHVHQWO\ D VWXGHQW" f
PAGE 99

+DYH \RX KDG DQ\ PDMRU LOOQHVVHV RYHU WKH SDVW PRQWKV" f
PAGE 100

+DV WKHUH EHHQ D FKDQJH LQ \RXU XVXDO SDWWHUQ RI VH[XDO DFWLYLW\ RYHU WKH SDVW PRQWKV" ,I VR GHVFULEH 7R ZKDW GR \RX DWWULEXWH WKLV FKDQJH" 'R \RX KDYH DQ\ LQIRUPDWLRQ WKDW \RX ZDQW WR DGG WR ZKDW ZHnYH DOUHDG\ WDONHG DERXW"

PAGE 101

$33(1',; % 68%-(&76n 727$/ 6&25(6 21 675(66 02'(5$725 $1' 5(&855(1&( 9$5,$%/(6 .H\ WR YDULDEOH QDPHV ,' 6XEMHFW LGHQWLILFDWLRQ FKDQJHG IRU SXEOLFDWLRQf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

PAGE 102

,' /(61 /(617 /(6,03 +$67 +$6,037 $6 666 /& +695(& +69326 1805(&,3 f O

PAGE 103

,' /(61 /(617 /(6,03 +$67 +$6,037 VVV /& +695(& +69326 1805(&,3 O O f $6

PAGE 104

,' /(61 /(617 /(6,03 +$67 +$6,037 $6 666 /& +695(& +69326 1805(&,3 m

PAGE 105

$33(1',; & 63($50$1 &255(/$7,21 &2()),&,(176 $1' 6,*1,),&$1&( /(9(/6 )25 5(3257(' 675(66 )5(48(1&< $1' 5(&855(1&( '(),1,7,216 5HFXUUHQFH 'HILQLWLRQV 5HSRUWHG 6WUHVV )UHTXHQF\ 0DMRU 1HJDWLYH /LIH (YHQWV +DVVOHV SDVW PRQWKV QB FRQFXUUHQW QM FRQFXUUHQW ML 7RWDO UHFRUGHG r f f f 'RFXPHQWHG E\ SRVLWLYH r FXOWXUH f f f $W OHDVW RQH GRFXPHQWHG r f f f 1RWH &RUUHODWLRQV DSSHDU DERYH VLJQLILFDQFH OHYHOV ZKLFK DUH LQ SDUHQWKHVHV r rr

PAGE 106

5()(5(1&(6 $GDP ( 'UHVVPDQ ( .DXIPDQ 5 + t 0HOQLFN / f $V\PSWRPDWLF YLUXV VKHGGLQJ DIWHU KHUSHV JHQLWDOLV $PHULFDQ -RXUQDO RI 2EVWHWULFV DQG *\QHFRORJ\ $ILIL $ $ t &ODUN 9 f &RPSXWHU DLGHG PXOWLYDULDWH DQDO\VLV %HOPRQW &$ :DGVZRUWK %DUWURS 5 : /XFNKXUVW ( /D]DUXV / .LORK / t 3HQQ\ 5 f 'HSUHVVHG O\PSKRF\WH IXQFWLRQ DIWHU EHUHDYHPHQW /DQFHW %HGHOO 5 *LRUGDQL % $UPRXU / 7DYRUPLQD t %ROO 7 f /LIH VWUHVV DQG WKH SV\FKRORJLFDO DQG PHGLFDO DGMXVWPHQW RI FKURQLFDOO\ LOO FKLOGUHQ -RXUQDO RI 3V\FKRVRPDWLF 5HVHDUFK %HUNPDQ / ) t 6\PH 6 / f 6RFLDO QHWZRUNV KRVW UHVLVWDQFH DQG PRUWDOLW\ $ QLQH\HDU IROORZXS VWXG\ RI $ODPHGD &RXQW\ UHVLGHQWV $PHULFDQ -RXUQDO RI (SLGHPLRORJ\ %HWWROL ( %UHZHU 3 0 2[WRE\ 0 $]LGL $ $ t *XLQDQ 0 ( f 7KH UROH RI WHPSHUDWXUH DQG VZDE PDWHULDOV LQ WKH UHFRYHU\ RI KHUSHV VLPSOH[ YLUXV IURP OHVLRQV -RXUQDO RI ,QIHFWLRXV 'LVHDVHV %LHUPDQ 6 0 f 7KH P\WK DQG UHDOLWLHV RI JHQLWDO KHUSHV 7KH -RXUQDO RI 0LQG DQG %HKDYLRU %LHUPDQ 6 0 f $ SRVVLEOH SV\FKRQHXURLPPXQRORJLF EDVLV IRU UHFXUUHQW JHQLWDO KHUSHV VLPSOH[ :HVWHUQ -RXUQDO RI 0HGLFLQH %LHUPDQ 6 0 f 5HFXUUHQW JHQLWDO KHUSHV VLPSOH[ LQIHFWLRQ $ WULYLDO GLVRUGHU $UFKLYHV RI 'HUPDWRORJ\ %ODQN + t %URG\ 0 : f 5HFXUUHQW KHUSHV VLPSOH[ $ SV\FKLDWULF DQG ODERUDWRU\ VWXG\ 3V\FKRVRPDWLF PHGLFLQH %O\WK : $ t +LOO f (VWDEOLVKPHQW PDLQWHQDQFH DQG FRQWURO RI KHUSHV VLPSOH[ YLUXV +69f ODWHQF\ ,Q %7 5RXVH DQG & /RSH[ (GVf ,PPXQRELRORJ\ RI KHUSHV VLPSOH[ YLUXV LQIHFWLRQ SS f %RFD 5DWRQ )/ &5& 3UHVV

PAGE 107

%UDGOH\ & f /LIH HYHQWV DQG WKH FRQWURO RI GLDEHWHV PHOOLWXV -RXUQDO RI 3V\FKRVRPDWLF 5HVHDUFK %UDPZHOO 6 7 0DVXGD 0 :DJQHU 1 1 t +ROPHV 7 + f 3V\FKRVRFLDO IDFWRUV LQ DWKOHWLF LQMXULHV -RXUQDO RI +XPDQ 6WUHVV %URZQ : %KUROFKDLQ 0 1 t +DUULV 7 f 6RFLDO FODVV DQG SV\FKLDWULF GLVWXUEDQFHV DPRQJ ZRPHQ LQ DQ XUEDQ SRSXODWLRQ 6RFLRORJ\ &DUWRQ & $ t .LOERXUQH ( f $FWLYDWLRQ RI ODWHQW KHUSHV VLPSOH[ E\ WULJHPLQDO VHQVRU\URRW VHFWLRQ 1HZ (QJODQG -RXUQDO RI 0HGLFLQH &OHDU\ 3 t $QJHO 5 f 7KH DQDO\VLV RI UHODWLRQVKLSV LQYROYLQJ GLFKRWRPRXV GHSHQGHQW YDULDEOHV -RXUQDO RI +HDOWK DQG 6RFLDO %HKDYLRU &OHDU\ 3 f $ FKHFNOLVW IRU OLIH HYHQW UHVHDUFK -RXUQDO RI 3V\FKRVRPDWLF 5HVHDUFK &RKHQ / + f /LIH FKDQJHV DQG WKH VHQVDWLRQ VHHNLQJ PRWLYH 3HUVRQDOLW\ DQG ,QGLYLGXDO 'LIIHUHQFHV &RKHQ 6 t +REHUPDQ + f 3RVLWLYH HYHQWV DQG VRFLDO VXSSRUWV DV EXIIHUV RI OLIH FKDQJH VWUHVV -RXUQDO RI $SSOLHG 6RFLDO 3V\FKRORJ\ &RKHQ 6 t :LOOV 7 $ f 6WUHVV VRFLDO VXSSRUW DQG WKH EXIIHULQJ K\SRWKHVLV 3V\FKRORJLFDO %XOOHWLQ &RQQROO\ f /LIH HYHQWV EHIRUH P\RFDUGLDO LQIDUFWLRQ -RXUQDO RI +XPDQ 6WUHVV &RUH\ / t +ROPHV . f *HQLWDO KHUSHV VLPSOH[ YLUXV LQIHFWLRQV &XUUHQW FRQFHSWV LQ GLDJQRVLV WKHUDS\ DQG SUHYHQWLRQ $QQDOV RI ,QWHUQDO 0HGLFLQH &RUH\ / 5HHYHV : & t +ROPHV . f &HOOXODU LPPXQH UHVSRQVH LQ JHQLWDO KHUSHV VLPSOH[ YLUXV LQIHFWLRQ 1HZ (QJODQG -RXUQDO RI 0HGLFLQH &UDLJ 7 t %URZQ : f *RDO IUXVWUDWLQJ DVSHFWV RI OLIH HYHQW VWUHVV LQ WKH HWLRORJ\ RI JDVWURLQWHVWLQDO GLVRUGHU -RXUQDO RI 3V\FKRVRPDWLF 5HVHDUFK &UHHG ) -XQHf /LIH HYHQWV DQG DSSHQGLFHFWRP\ /DQFHW f &UHHG ) f ,QYLWHG UHYLHZ /LIH HYHQWV DQG SK\VLFDO LOOQHVV -RXUQDO RI 3V\FKRVRPDWLF 5HVHDUFK

PAGE 108

GH $UDXMR 9DQ $UVGHO 3 3 +ROPHV + t 'XGOH\ / f /LIH FKDQJH FRSLQJ DELOLW\ DQG FKURQLF LQWULQVLF DVWKPD -RXUQDO RI 3V\FKRVRPDWLF 5HVHDUFK 'H/RQJLV $ &R\QH & 'DNRI )RONPDQ 6 t /D]DUXV 5 6 f 5HODWLRQVKLS RI GDLO\ KDVVOHV XSOLIWV DQG PDMRU OLIH HYHQWV WR KHDOWK VWDWXV +HDOWK 3V\FKRORJ\ 'RKUHQZHQG % 3 f 3UREOHPV LQ GHILQLQJ DQG VDPSOLQJ WKH UHOHYDQW SRSXODWLRQ RI VWUHVVIXO OLIH HYHQWV ,Q %6 'RKUHQZHQG DQG %3 'RKUHQZHQG (GVf 6WUHVVIXO OLIH HYHQWV 7KHLU QDWXUH DQG HIIHFWV SS f 1HZ
PAGE 109

*UHHQ 0 7 5RVERURXJK 3 t 'XQNHO ( & f ,Q YLYR UHDFWLYDWLRQ RI KHUSHV VLPSOH[ YLUXV LQ UDEELW WULJHPLQDO JDQJOLD (OHFWURGH PRGHO ,QIHFWLRQ DQG ,PPXQLW\ f +DOO 1 5 t *ROGVWHLQ $ / f 1HXURWUDQVPLWWHUV DQG WKH LPPXQH V\VWHP ,Q 5 $GHU (Gf 3V\FKRQHXUR LPPXQRORJ\ SS f 1HZ
PAGE 110

+XUVW 0 : -HQNLQV & t 5RVH 5 0 f 7KH UHODWLRQ RI SV\FKRORJLFDO VWUHVV WR RQVHW RI PHGLFDO LOOQHVV $QQXDO 5HYLHZ RI 0HGLFLQH -DFREV 7 t &KDUOHV ( f /LIH HYHQWV DQG WKH RFFXUUHQFH RI FDQFHU LQ FKLOGUHQ 3V\FKRVRPDWLF 0HGLFLQH -DFREVRQ $ 0 5DQG / t +DXVHU 6 7 f 3V\FKRORJLF VWUHVV DQG JO\FHPLF FRQWURO $ FRPSDULVRQ RI SDWLHQWV ZLWK DQG ZLWKRXW SUROLIHUDWLYH GLDEHWLF UHWLQRSDWK\ 3V\FKRVRPDWLF 0HGLFLQH -DQLFNL 0 3 f 5HFXUUHQW KHUSHV ODELDOLV DQG UHFXUUHQW DSKWKRXV XOFHUDWLRQV 3V\FKRORJLFDO FRPSRQHQWV 3V\FKRWKHUDS\ 3V\FKRPDWLFV -HQNLQV & +XUVW 0 : t 5RVH 5 0 f /LIH FKDQJHV 'R SHRSOH UHDOO\ UHPHPEHU" $UFKLYHV RI *HQHUDO 3V\FKLDWU\ -RKQVRQ + t 6DUDVRQ f /LIH VWUHVV GHSUHVVLRQ DQG DQ[LHW\ ,QWHUQDOH[WHUQDO FRQWURO DV D PRGHUDWRU YDULDEOH -RXUQDO RI 3V\FKRVRPDWLF 5HVHDUFK -RKQVRQ + t 6DUDVRQ f /LIH VWUHVV UHVHDUFK :KHUH KDYH ZH EHHQf§ZKHUH DUH ZH JRLQJ" 8QSXEOLVKHG PDQXVFULSW 8QLYHUVLW\ RI )ORULGD 'HSDUWPHQW RI &OLQLFDO 3V\FKRORJ\ *DLQHVYLOOH )/ -RKQVRQ + 6DUDVRQ t 6LHJHO 0 f $URXVDO VHHNLQJ DV D PRGHUDWRU RI OLIH VWUHVV 3HUFHSWXDO DQG 0RWRU 6NLOOV .DQQHU $ &R\QH & 6FKDHIHU & t /D]DUXV 5 6 f &RPSDULVRQ RI WZR PRGHV RI PHDVXUHPHQW 'DLO\ KDVVOHV DQG XSOLIWV YHUVXV PDMRU OLIH HYHQWV -RXUQDO RI %HKDYLRUDO 0HGLFLQH .DVO 6 9 (YDQV $ 6 -DPHV & t 1LHGHUPDQ & f 3V\FKRVRFLDO ULVN IDFWRUV LQ WKH GHYHORSPHQW RI LQIHFWLRXV PRQRQXFOHRVLV 3V\FKRVRPDWLF 0HGLFLQH .DWFKHU $ + %ULJKWPDQ 9 /XERUVN\ / t 6KLS f 3UHGLFWLRQ RI WKH LQFLGHQFH RI UHFXUUHQW KHUSHV ODELDOLV DQG V\VWHPLF LOOQHVV IURP SV\FKRORJLFDO PHDVXUHPHQWV -RXUQDO RI 'HQWDO 5HVHDUFK .HGGLH ) 0 5HHYHV 5 % t (SVWHLQ 1 1 f +HUSHV VLPSOH[ IROORZLQJ DUWLILFLDO IHYHU WKHUDS\ -RXUQDO RI WKH $PHULFDQ 0HGLFDO $VVRFLDWLRQ

PAGE 111

.HUQ $ f 5HFXUUHQW KHUSHV VLPSOH[ $ VLPSOH WUHDWPHQW &RQVXOWDQW .LHFROW*ODVHU 5LFNHU *HRUJH 0HVVLFN 6SHLFKHU & ( *DUQHD : t *ODVHU 5 f 8ULQDU\ FRUWLVRO OHYHOV FHOOXODU LPPXQRFRPSHWHQF\ DQG ORQHOLQHVV 3V\FKRVRPDWLF 0HGLFLQH .LUFKQHU + 6FKZHQWHFN 0 1RUWKRII + t 6FKRSI ( f 'HIHFWLYH KL YLWUR O\PSKRSUROLWHUDWLYH UHVSRQVHV WR KHUSHV VLPSOH[ YLUXV LQ SDWLHQWV ZLWK IUHTXHQWO\ UHFXUULQJ KHUSHV LQIHFWLRQV GXULQJ WKH GLVHDVHIUHH LQWHUYDO &OLQLFDO ,PPXQRORJ\ DQG ,PPXQRSDWKRORJ\ .REDVD 6 & f 6WUHVVIXO OLIH HYHQWV SHUVRQDOLW\ DQG KHDOWK $Q LQTXLU\ LQWR KDUGLQHVV -RXUQDO RI 3HUVRQDOLW\ DQG 6RFLDO 3V\FKRORJ\ .REDVD 6 & 0DGGL 6 5 t &RXULQJWRQ 6 f 3HUVRQDOLW\ DQG FRQVWLWXWLRQ DV PHGLDWRUV LQ WKH VWUHVVLOOQHVV UHODWLRQVKLS -RXUQDO RI +HDOWK DQG 6RFLDO %HKDYLRU /DLEVRQ 3 5 t .LEULFN 6 f 5HDFWLYDWLRQ RI KHUSHWLF NHUDWLWLV E\ HSLQHSKULQH LQ UDEELW $UFKLYHV RI 2SWKDPRORJ\ /DL 1 $KXMD 5 & t 0DGKXNDU $ f /LIH HYHQWV LQ K\SHUWHQVLYH SDWLHQWV -RXUQDO RI 3V\FKRVRPDWLF 5HVHDUFK /HYHQVRQ + +LUVFKIHOG 0 / +LUVFKIHOG $ t ']XED\ % f -RXUQDO RI +XPDQ 6WUHVV /LDR : f 3V\FKRORJLFDO VWUHVV KHDOWK VWDWXV DQG KHDOWK EHKDYLRU $SSOLFDWLRQ RI WKH OLIH FKDQJHV FRQFHSW 3V\FKRORJLFDO 5HSRUWV /LQ 1 (QVHO : 0 6LPHRQH 5 6 t .XR : f 6RFLDO VXSSRUW VWUHVVIXO OLIH HYHQWV DQG LOOQHVV $ PRGHO DQG DQ HPSLULFDO WHVW -RXUQDO RI +HDOWK DQG 6RFLDO %HKDYLRU /LQQ 0 : f ,Q GHIHQVH RI VWDWLVWLFDO VLJQLILFDQFH ZLWK PXOWLYDULDWH DQDO\VLV 'LDEHWHV &DUH /LQQ 0 : /LQQ % 6 6N\OHU 6 t -HQVHQ f 6WUHVV DQG LPPXQH IXQFWLRQ LQ GLDEHWHV PHOOLWXV &OLQLFDO LPPXQRORJ\ DQG LPPXQRSDWKRORJ\ /RFNH 6 ( f 6WUHVV DGDSWDWLRQ DQG LPPXQLW\ 6WXGLHV LQ KXPDQV *HQHUDO +RVSLWDO 3V\FKLDWU\

PAGE 112

/XERUVN\ / 0LQW] %ULJKWPDQ 9 t .DWFKHU $ + f +HUSHV VLPSOH[ DQG PRRGV $ ORQJLWXGLQDO VWXG\ -RXUQDO RI 3V\FKRVRPDWLF 5HVHDUFK /XE\ ( t .OLQJH 9 f *HQLWDO KHUSHV $ SHUYDVLYH SV\FKRVRFLDO GLVRUGHU $UFKLYHV RI 'HUPDWRORT\ 0DQXFN 6 % +LQULFKVHQ t 5RVV ( f /LIH VWUHVV ORFXV RI FRQWURO DQG WUHDWPHQW VHHNLQJ 3V\FKRORJLFDO 5HSRUWV 0DU[ 0 % *DUULW\ 5 ) t %RZHUV ) 5 f 7KH LQIOXHQFH RI UHFHQW OLIH H[SHULHQFH RQ WKH KHDOWK RI FROOHJH IUHVKPHQ -RXUQDO RI 3V\FKRVRPDWLF 5HVHDUFK 0F'RZHOO 0 f +\SQRVLV LQ GHUPDWRORJ\ ,Q -0 6FKQHFN (Gf +\SQRVLV LQ PRGHUQ PHGLFLQH QG HG SS f 6SULQJILHOG ,/ &KDUOHV & 7KRPDV 0F)DUODQH $ + 1RUPDQ 5 6WUHLQHU / t 5R\ 5 f 7KH SURFHVV RI VRFLDO VWUHVV 6WDEOH UHFLSURFDO DQG PHGLDWLQJ UHODWLRQVKLSV -RXUQDO RI +HDOWK DQG 6RFLDO %HKDYLRU 0HFKDQLF f 6RPH SUREOHPV LQ WKH PHDVXUHPHQW RI VWUHVV DQG VRFLDO UHDGMXVWPHQW -RXUQDO RI +XPDQ 6WUHVV 0HKUDELDQ $ f &KDUDFWHULVWLF LQGLYLGXDO UHDFWLRQV WR SUHIHUUHG DQG XQSUHIHUUHG HQYLURQPHQWV -RXUQDO RI 3HUVRQDO LW\ 0HKUDELDQ $ t 5XVVHOO $ f $Q DSSURDFK WR HQYLURQPHQWDO SV\FKRORJ\ SS f &DPEULGJH 0$ 0,7 3UHVV 0HUW] t &RUH\ / f *HQLWDO KHUSHV VLPSOH[ YLUXV LQIHFWLRQV LQ DGXOWV 8URORJLF &OLQLFV RI 1RUWK $PHULFD f 0H\HU 5 t +DJJHUW\ 5 f 6WUHSWRFRFFDO LQIHFWLRQ LQ IDPLOLHV )DFWRUV LQ WKH RQVHW RI VRPDWLF LOOQHVV LQ FKLOGUHQ 3HGLDWULFV 0H\HUV )ORXUQR\ 1 t 7KRPDV ( f ,QIHFWLRQ ZLWK KHUSHV VLPSOH[ YLUXV DQG FHOOPHGLDWHG LPPXQLW\ DIWHU PDUURZ WUDQVSODQW -RXUQDO RI ,QIHFWLRXV 'LVHDVHV 0LOOHU 30 ,QJKDP -* t 'DYLGVRQ 6 f /LIH HYHQWV V\PSWRPV DQG VRFLDO VXSSRUW -RXUQDO RI 3V\FKRVRPDWLF 5HVHDUFK 0LQWHU 5 ( t .LPEDOO & 3 f /LIH HYHQWV DQG LOOQHVV RQVHW $ UHYLHZ 3V\FKRVRPDWLFV

PAGE 113

0RQURH 6 0 Df $VVHVVPHQW RI OLIH HYHQWV 5HWURVSHFWLYH YV FRQFXUUHQW VWUDWHJLHV $UFKLYHV RI *HQHUDO 3V\FKLDWU\ 0RQURH 6 0 Ef /LIH HYHQWV DVVHVVPHQW &XUUHQW SUDFWLFHV HPHUJLQJ WUHQGV &OLQLFDO 3V\FKRORJ\ 5HYLHZ 0XHOOHU 3 (GZDUGV : t
PAGE 114

3DVVHU 0 : t 6HHVH 0 'HFHPEHUf /LIH VWUHVV DQG DWKOHWLF LQMXU\ ([DPLQDWLRQ RI SRVLWLYH YHUVXV QHJDWLYH HYHQWV DQG WKUHH PRGHUDWRU YDULDEOHV -RXUQDO RI +XPDQ 6WUHVV f 3D\NHO ( 6 f 0HWKRGRORJLFDO DVSHFWV RI OLIH HYHQWV UHVHDUFK -RXUQDO RI 3V\FKRVRPDWLF 5HVHDUFK 3D]LQ $UPVWURQJ $ /DP 0 7 7DUU & -DQHWWD 3 t +R 0 f 3UHYHQWLRQ RI UHDFWLYDWHG KHUSHV VLPSOH[ LQIHFWLRQ E\ KXPDQ OHXNRF\WH LQWHUIHURQ DIWHU RSHUDWLRQ RQ WKH WULJHPLQDO URRW 1HZ (QJODQG -RXUQDO RI 0HGLFLQH 3ODXW 6 0 t )ULHGPDQ 6 % f 3V\FKRVRFLDO IDFWRUV LQ LQIHFWLRXV GLVHDVH ,Q 5( $GHU (Gf 3V\FKRQHXURLPPXQRORJ\ SSf 1HZ
PAGE 115

5RWK / t +ROPHV 6 f ,QIOXHQFH RI SK\VLFDO ILWQHVV LQ GHWHUPLQLQJ WKH LPSDFW RI VWUHVVIXO OLIH HYHQWV RQ SK\VLFDO DQG SV\FKRORJLF KHDOWK 3V\FKRVRPDWLF 0HGLFLQH 5RWWHU % f *HQHUDOL]HG H[SHFWDQFLHV IRU LQWHUQDO YV H[WHUQDO FRQWURO RI UHLQIRUFHPHQW 3V\FKRORJLFDO 0RQRJUDSKV 6HULDO 1R f 6DUDVRQ -RKQVRQ + t 6LHJHO 0 f 'HYHORSPHQW RI WKH OLIH H[SHULHQFHV VXUYH\ -RXUQDO RI &RQVXOWLQJ DQG &OLQLFDO 3V\FKRORJ\ 6DUDVRQ /HYLQH + 0 %DVKDP 5 % t 6DUDVRQ % 5 f $VVHVVLQJ VRFLDO VXSSRUW 7KH VRFLDO VXSSRUW TXHVWLRQQDLUH 15 5HSRUW 1R &f $UOLQJWRQ 9$ 2IILFH RI 1DYDO 5HVHDUFK 6DUDVRQ /HYLQH + 0 t 6DUDVRQ % 5 f $VVHVVLQJ WKH LPSDFW RI OLIH FKDQJHV ,Q 7 0LOOQ & *UHHQ DQG 5 0HDJKHU (GVf +DQGERRN RI FOLQLFDO KHDOWK SV\FKRORJ\ SS f 1HZ
PAGE 116

6PLWK 5 ( -RKQVRQ + t 6DUDVRQ f /LIH FKDQJH WKH VHQVDWLRQ VHHNLQJ PRWLYH DQG SV\FKRORJLFDO GLVWUHVV -RXUQDO RI &RQVXOWLQJ DQG &OLQLFDO 3V\FKRORJ\ 6WHYHQV f /DWHQW KHUSHV VLPSOH[ YLUXV DQG WKH QHUYRXV V\VWHP &XUUHQW 7RSLFV LQ 0LFURELRORJ\ DQG ,PPXQRORJ\ 6WUDXV 6 ( 7DNLII + ( 6HLGOLQ 0 %DFKUDFK 6 /LQQLQJHU / 'L*LRYDQQD :HVWHUQ $ 6PLWK + $ /HKUPDQ 6 1 &UHDJK.LUN 7 t $LOLQJ : f 6XSSUHVVLRQ RI IUHTXHQWO\ UHFXUULQJ JHQLWDO KHUSHV $ SODFHERFRQWUROOHG GRXEOHEOLQG WULDO RI RUDO DF\FORYLU 1HZ (QJODQG -RXUQDO RI 0HGLFLQH 7D\ORU % f 7KH SV\FKRORJLFDO DQG EHKDYLRUDO HIIHFWV RI JHQLWDO KHUSHV LQ ZRPHQ +LJK UHFXUUHUV YV ORZ UHFXUUHUV 'RFWRUDO GLVVHUWDWLRQ 8QLYHUVLW\ RI :DVKLQJWRQ f 'LVVHUWDWLRQ $EVWUDFWV ,QWHUQDWLRQDO %f 8QLYHUVLW\ 0LFURILOPV 1R f 7HVVOHU 5 0HFKDQLF t 'LPRQG 0 f 7KH HIIHFW RI SV\FKRORJLFDO GLVWUHVV RQ SK\VLFLDQ XWLOL]DWLRQ $ SURVSHFWLYH VWXG\ -RXUQDO RI +HDOWK DQG 6RFLDO %HKDYLRU 7KHRUHOO 7 t 5DKH 5 + f 3V\FKRVRFLDO IDFWRUV DQG P\RFDUGLDO LQIDUFWLRQ $Q LQSDWLHQW VWXG\ LQ 6ZHGHQ -RXUQDO RI 3V\FKRVRPDWLF 5HVHDUFK 7KRLWV 3 $ f &RQFHSWXDO PHWKRGRORJLFDO DQG WKHRUHWLFDO SUREOHPV LQ VWXG\LQJ VRFLDO VXSSRUW DV D EXIIHU DJDLQVW OLIH VWUHVV -RXUQDO RI +HDOWK DQG 6RFLDO %HKDYLRU 7KXUORZ + f ,OOQHVV LQ UHODWLRQ WR OLIH VLWXDWLRQ DQG VLFN UROH WHQGHQF\ -RXUQDO RI 3V\FKRVRPDWLF 5HVHDUFK 7RWPDQ 5 .LII 5HHG 6 ( t &UDLJ : f 3UHGLFWLQJ H[SHULPHQWDO FROGV LQ YROXQWHHUV IURP GLIIHUHQW PHDVXUHV RI UHFHQW OLIH VWUHVV -RXUQDO RI 3V\FKRVRPDWLF 5HVHDUFK 7RYHV & 6FKLOO 7 t 5DPDQDLDK 1 f 6H[ GLIIHUHQFHV LQWHUQDOH[WHUQDO FRQWURO DQG YXOQHUDELOLW\ WR OLIH VWUHVV 3V\FKRORJLFDO 5HSRUWV 8OOPDQ 0 f +HUSHV VLPSOH[ DQG VHFRQG GHJUHH EXUQ LQGXFHG XQGHU K\SQRVLV $PHULFDQ -RXUQDO RI 3V\FKLDWU\

PAGE 117

QR 9LQRNXU $ t 6HO]HU 0 / f 'HVLUDEOH YHUVXV XQGHVLUDEOH OLIH HYHQWV 7KHLU UHODWLRQVKLS WR VWUHVV DQG PHQWDO GLVWUHVV -RXUQDO RI 3HUVRQDOLW\ DQG 6RFLDO 3V\FKRORTY :DWVRQ % f 7KH UHODWLRQVKLS RI JHQLWDO KHUSHV DQG OLIH VWUHVV DV PRGHUDWHG E\ ORFXV RI FRQWURO DQG VRFLDO VXSSRUW 'RFWRUDO GLVVHUWDWLRQ 8QLYHUVLW\ RI 6RXWKHUQ &DOLIRUQLD f 'LVVHUWDWLRQ $EVWUDFWV ,QWHUQDWLRQDO $f :LFNHWW : + f +HUSHV &DXVH DQG FRQWURO 1HZ
PAGE 118

%,2*5$3+,&$/ 6.(7&+ (PLO\ )UDQFN +RRQ ZDV ERUQ DQG JUHZ XS ZLWK KHU SDUHQWV DQG RQO\ VLEOLQJ D \RXQJHU VLVWHU LQ &RQQHFWLFXW $IWHU FRPSOHWLQJ KLJK VFKRRO LQ &RQQHFWLFXW VKH DWWHQGHG *ULQQHOO &ROOHJH LQ ,RZD ZKHUH VKH UHFHLYHG D %$ GHJUHH LQ SV\FKRORJ\ (PLO\ LQYROYHG KHUVHOI IRU \HDUV WHDFKLQJ HGXFDEOH PHQWDOO\ UHWDUGHG FKLOGUHQ LQ 1HZ
PAGE 119

FUHGLW KDYH JURZQ IURP WKLV EHJLQQLQJ :KLOH VHUYLQJ DV D SDUWWLPH PHPEHU RI WKH 3V\FKRORJ\ 'HSDUWPHQW DW WKH 1RYD 6FRWLD 3URYLQFLDO +RVSLWDO (PLO\ HDUQHG KHU PDVWHUn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f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

PAGE 120

, FHUWLI\ WKDW KDYH UHDG WKLV VWXG\ DQG WKDW LQ P\ RSLQLRQ LW FRQIRUPV WR DFFHSWDEOH VWDQGDUGV RI VFKRODUO\ SUHVHQWDWLRQ DQG LV IXOO\ DGHTXDWH LQ VFRSH DQG TXDOLW\ DV D GLVVHUWDWLRQ IRU WKH GHJUHH RI 'RFWRU RI 3KLORVRSK\ ? &? 8 -DPHV + -RKQVRQ &KDLUPDQ $VVRFLDWH 3URIHVVRU RI &OLQLFDO 3V\FKRORJ\ FHUWLI\ WKDW KDYH UHDG WKLV VWXG\ DQG WKDW LQ P\ RSLQLRQ LW FRQIRUPV WR DFFHSWDEOH VWDQGDUGV RI VFKRODUO\ SUHVHQWDWLRQ DQG LV IXOO\ DGHTXDWH LQ VFRSH DQG TXDOLW\ DV D GLVVHUWDWLRQ IRU WKH GHJUHH RI 'RFWRU RI 3KLORVRSK\ 5DQG\ &DUWH $VVRFLDWH 3URIHVVRU RI 6WDWLVWLFV FHUWLI\ WKDW KDYH UHDG WKLV VWXG\ DQG WKDW LQ P\ RSLQLRQ LW FRQIRUPV WR DFFHSWDEOH VWDQGDUGV RI VFKRODUO\ SUHVHQWDWLRQ DQG LV IXOO\ DGHTXDWH LQ VFRSH DQG TXDOLW\ DV D GLVVHUWDWLRQ IRU WKH GHJUHH RI 'RFWRU RI 3KLORVRSK\ O8XA &E FANM--/ (LOHHQ % )HQQHOO $VVRFLDWH 3URIHVVRU RI &OLQLFDO 3V\FKRORJ\ FHUWLI\ WKDW KDYH UHDG WKLV VWXG\ DQG WKDW LQ P\ RSLQLRQ LW FRQIRUPV WR DFFHSWDEOH VWDQGDUGV RI VFKRODUO\ SUHVHQWDWLRQ DQG LV IXOO\ DGHTXDWH LQ VFRSH DQG TXDOLW\ DV D GLVVHUWDWLRQ IRU WKH GHJUHH RI 'RFWRU RI 3KLORVRSK\ 787XDM" J 1DQF\ 1RUYGMO $VVLVWDQW 3URIHVVRU RI &OLQLFDO 3V\FKRORJ\

PAGE 121

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MA Ln? LQDQ 3QO LSKS QU 8QD X-/U n'HDQ &ROOHJH RI +HDOWK 5HODWHG 3URIHVVLRQV 'HDQ *UDGXDWH 6FKRRO

PAGE 122

81,9(56,7< 2) )/25,'$