|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
PERSONALITY TRAITS AS RISK FACTORS FOR OCCUPATIONAL INJURY IN HEALTH
HILARY STEVENS MORGAN
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
O 2007 Hilary Stevens Morgan
To my parents who always stressed the value of education and instilled in their children a belief
that anything was possible
I would like to thank my committee chair, Dr. Nancy Nivison Menzel, for her guidance,
encouragement, and patience throughout this research study. I appreciate her knowledge of
occupational health nursing and her commitment to provide leadership for this study despite
moving over 2,000 miles westward.
I gratefully acknowledge and extend my appreciation to the members of my committee, Jo
Snider, PhD, Jiunn-Jye Sheu, PhD, and James V. Jessup, PhD. Each of them provided unique
talents, time, and moral support during this study. I would like to thank Dr. Snider for her
support during my struggles understanding psychology, Dr. Sheu for his patience in dealing with
my epidemiology and study design questions and Dr. Jessup for his calm and encouragement in
my moments of doubt.
I also want to thank Seaborn Hunt, MD for his support of my returning to school. I know
my absences in the office for school activities presented an additional burden on him, but he
never wavered in his encouragement of my endeavors. The same appreciation is extended to
Paula Murphy, Kathy Sandor, Ronnie Maxim, Vickie Hall, Milly Wilkinson and Renee' Brown
who tolerated, with good graces, my vacillating moods throughout this long process. I also want
to extend thanks to "my friend" Alice Carlisle. We began this journey as PhD students together
and it would not have been as successful or as enj oyable without her companionship, friendship
and advice over these last four years. Lastly, a special thanks to my sister, Lesley Morgan, PhD,
RN, who paved the road before me and provide much insight, guidance and encouragement to
get me to the end of the j ourney successfully.
TABLE OF CONTENTS
ACKNOWLEDGMENT S .............. ...............4.....
LIST OF TABLES ................ ...............8............ ....
LI ST OF FIGURE S .............. ...............9.....
List of Key Terms ................ ...............10................
AB S TRAC T ............._. .......... ..............._ 13...
1 INTRODUCTION ................. ...............14......... .....
Background of the Problem ................. ...............14......... ....
Costs of Safety/Injury ................. ...............15...............
Health Care W workers ................ .. .......... ............1
History of Occupational Health and Safety ..........._._ ........... ........._..........1
The Role of NIOSH and OSHA ................. ...............17.............
Epidemiology .............. .. ......... ..............1
Theories on Accidents and Safety .............. ...............19....
The Safe Environment ................. ...............22........... ....
Personality Theories .............. ........ .... .... .......2
Biological and genetic theories on personality development ................. ................23
Environmental theories on personality ................. ...............25................
Situation vs. person .............. ...............26....
Development of traits ................ .............. ........ ......... ........ .....26
Industrial/Organizational Psychology .............. ...............27....
Ethical Considerations in Personality Testing ......___ ..... ... ._ ........__.........3
Purpose of the Study ............_ ..... ..__ ...............33..
Significance of the Study ............_ ..... ..__ ...............33..
Theoretical Framework............... ...............3
Social Ecology Theory .................. ...............34..
The PRECEDE-PROCEED Model .............. ...............35....
Epidemiological Triad ................. ...............36........_......
Hyp others e s................. ...............37......... ....
Variables ....._. ................ ........_.. .........37
2 REVIEW OF LITERATURE ..........._.__........... ...............39.....
Agent Factors ................. ...............39......... ......
Environmental Factors ................. ...............41......... ......
Host Factors ................ .. .....___ ...............44......
Personality, Accidents and Injuries .............. ...............46....
The Five Factor Model in the Workplace ........._._._....._. .. ...............54..
Summary of Literature Review ........_.............. .............._........ ......._...56
3 METHODS .............. ...............58....
Research Design .............. ...............58....
Research Setting .............. ...............59....
Sam ple .............. ...............59....
Sample Criteria ............ _...... ._ ...............59....
M measures .............. ....._ ...............60....
Personality Inventory............... ...............6
Job Relative Ri sk ............... ...............62..
Demographic Information .............. ...............62....
Procedures ............... ....._ ...............63....
Data Collection Procedures .............. .......... ..............6
Procedures for the Protection of Human Subj ects ...._._._.. ............. ......_... ........6
Data Analysis............... ...............68
Power Analysis ........._...... ...............69..._.._. ......
4 ANALYSIS AND RESULTS................ ...............70
Demographic Statistics .............. ...............70....
Personality Statistics ................. ...............71...............
Research Question .............. ...............72....
5 CONCLUSIONS AND RECOMMENDATIONS .............. ...............77....
Discussion of Findings .............. ...............77....
Hyp others e s.........._...._ ......... ...............79....
Limitations of Study .............. ...............80....
Study Design Limitations .........._...._ ...............80.......__......
Statistical Analysis Limitation .............. ...............82....
Strengths of the Study .............. ...............82....
Conclusions............... .. .. .. ..................8
Recommendations for Further Research .............. ...............84....
Implications for Clinical Practice .............. ...............85....
A JOB RISK QUESTIONNAIRE .........._...._ ......... ...............86.....
B DEMOGRAPHIC QUESTIONNAIRE............... .............8
LIST OF REFERENCES ............._.. ...............88....._.........
BIOGRAPHICAL SKETCH .............. ...............98....
LIST OF TABLES
4-1 Demographic statistics for job classification, age and tenure for the overall sample
and by group (injured vs. non-injured) .............. ...............74....
4-2 Means and standard deviation of the personality domains for the overall sample and
by group (inj ured vs. non-inj ured) ................ ...............74........... ..
4-3 Means and standard deviation for the personality sub facet scores for the overall
sample and by group (injured vs. non-injured) ....._._.__ ............ ... ...._..........7
4-4 Classifieation by group (injured vs. non-injured) by excitement-seeking,
impulsiveness, angry hostility and compliance (Step 2) after controlling for job
classification, age and tenure (Step 1) .............. ...............76....
4-5 Logistic regression on excitement seeking, impulsiveness, compliance and angry
hostility predicting group membership (injured vs. non-injured) after controlling for
age, tenure and j ob classification (nurse vs. UAP) ................ ............... ......... ...76
LIST OF FIGURES
1-1 Planning model to prevent occupational injury .............. ...............37....
1-2 The PROCEED-PRECEDE Model ........._._. ...._. ...............38..
1-3 Epidemiological triad............... ...............38.
LIST OF KEY TERMS
"That occurrence in a sequence of events that produces
unintended injury, death or property damage. Accident
refers to the event, not the result of the event" (National
Safety Council [NSC], 2005).
"The quality of being agreeable or pleasing, that quality
which gives satisfaction or moderate pleasure to the mind
of senses" (Merriam-Webster's Online, 2005). A
"dimension of interpersonal tendencies; fundamentally
altruistic" (Costa & McCrae, 1992, p. 15).
"A deep seated ill will. Conflict, opposition, or resistance
in thought or principle" (Merriam-Webster' s Online, 2005).
"The tendency to experience anger and related states such
as frustration and bitterness" (Costa & McCrae, 1992, p.
"The tendency to defer to others, to inhibit aggression and
to forgive and forget" (Costa & McCrae, 1992, p. 18).
"The disposition to defer to others" (Merriam-Webster' s
"The appetite for the thrills of bright colors and noisy
settings" (Costa & McCrae, 1992, p. 17).
"The preference for attending to the outer world of
obj ective events with an emphasis upon active involvement
in the environment. The extrovert is sociable, lively,
novelty-seeking, carefree and emotionally expressive"
(Morris, quoted in Hansen, 1988). Extraverts like
"excitement and stimulation and tend to be cheerful in
disposition" (Costa & McCrae, 1992, p. 15).
"A condition or physical situation with a potential for an
undesirable consequence, such as loss of life or limb"
Any person, including a student or trainee, whose activities
involve contact with patients or with blood or other body
fluids from patients in a health care setting (Centers for
Disease Control [CDC], 2005).
Health Care Worker
"The quality of being impulsive. Acting under stress of
emotion or spirit of the moment; acting without
deliberation" (Merriam-Webster's Online, 2005). "The
inability to control cravings and urges" (Costa & McCrae,
1992, p. 16).
"Physical harm or damage to the body resulting from an
exchange, usually acute, of mechanical, chemical, thermal,
or other environmental energy that exceeds the body's
tolerance" (NSC, 2005).
Someone who has specific education and training in
nursing and is licensed by a state professional board.
Licensed nurses include both registered nurses (RN) and
licensed practical nurses (LPN).
Individuals who are educated to diagnose and treat medical
conditions. Medical providers include physicians, nurse
practitioners, nurse anesthetists, nurse midwives and
The tendency to experience negative emotions such as fear,
anger, guilt, and sadness. Individuals high in neuroticism
also are less able to cope with life events and to control
their impulses; emotional stability (Roberts & Hogan,
Any abnormal condition or disorder, other than one
resulting from an occupational injury, caused by exposure
to environmental factors associated with employment. It
includes acute and chronic illnesses or diseases that may be
caused by inhalation, absorption, ingestion, or direct
contact. (NSC, 2005)
"Any such injury such as a cut, fracture, sprain,
amputation, etc., which results from a work accident or
from a single instantaneous exposure in the work
environment" (NSC, 2005).
The potential for realization of unwanted, adverse
consequences to human life, health property or to the
environment; estimate of risk is usually based on the
expected value of the conditional probability of the event
Source of injury
Unlicensed Assistive Personnel
occurring times the consequence of the event given that it
has occurred. (Merriam-Webster's Online, 2005)
"The state of being relatively free from hazards that are
likely to cause harm, injury or property damage" (NSC,
"The principal obj ect such as tool, machine, or equipment
involved in the accident and is usually the obj ect inflicting
injury or property damage. Also called agency or agent"
Individuals who have specific training in particular clinical
area such as the laboratory, radiology, surgery and
pharmacy. They assist licensed personnel with patient care
procedures. Technicians include clinical laboratory
technicians, radiology technicians, surgery technicians,
cardiovascular technicians, emergency medical technicians
and paramedics, diagnostic medical sonographers,
respiratory technicians and pharmacy technicians.
Individuals trained in methods of treatment and
rehabilitation other than the use of drugs or surgery.
Therapists include respiratory therapists, occupational
therapists and physical therapists.
"The preferred term for accidental injury in the public
health community; it refers to the result of an accident"
Unlicensed individuals who are trained to function in an
assistive role to the licensed nurse or other health
professional in the provision of patient/client activities as
delegated (American Nurses Association, [ANA], 1997).
UAPs include nursing assistants (NAs), health care
assistants, patient care technicians (PCTs), medical
assistants (MAs) and phlebotomists. Nurse interns (nursing
students working outside their normal clinical rotations)
also fall under this category.
(Including occupational illnesses) are defined as: those that
arise out of and in the course of gainful employment
regardless of where the accident or exposure occurs.
Excluded are work injuries to private household workers
and injuries occurring in connection with farm chores that
are classified as home injuries (NSC, 2005).
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
PERSONALITY TRAITS AS RISK FACTORS FOR OCCUPATIONAL INJURY IN HEALTH
Hilary Stevens Morgan
Chair: Nancy Nivison Menzel
Millions of dollars are spent each year as a result of injuries suffered by employees at
work. Health care workers, particularly nursing personnel, are among the groups at greatest risk
for injury. Research has shown individuals possessing certain personality traits to be more
vulnerable to injury. Individuals who are considered extraverted, neurotic and disagreeable are
hypothesized to be more at risk for injury although these findings have not been substantiated in
health care workers. Therefore, the purpose of this study was to identify certain personality trait
risk factors for occupational injuries in health care workers. This study used a case comparison
design to test the hypotheses that group membership (injured vs. non-injured workers) could be
predicted by the personality traits of excitement seeking, impulsiveness, angry hostility and
compliance after controlling for age, tenure and j ob classification.
Seventy two subjects (29 cases and 43 controls) were enrolled in the study. Both groups
were given the NEO Personality Inventory- Revised (NEO PI-R). Although the characteristics
of the groups were similar, group membership (injured vs. non-injured) was not predicted by the
personality traits of angry hostility, excitement seeking, impulsiveness and compliance (after
controlling for age, tenure and job classification). The final model improved group prediction
but not significantly.
Background of the Problem
The workforce of the United States in 2005 was estimated to be approximately 140 million
adult men and women between the ages of 18 to 65 years. Fully 75% of adult men and between
30 and 50% of adult women are part of the workforce at any time (Bureau of Labor Statistics
[BLS], 2005). Most spend between 40 and 50% of their waking hours at work. Each year
millions suffer injuries at work; several thousand die or sustain permanent disability or disease.
Costs to the economy are staggering. Estimated direct and indirect costs resulting from these
injuries and diseases are $13 1 billion annually or nearly 3% of the gross domestic product
(National Safety Council [NSC], 2003). Although it is impossible for any job to be completely
free from risk of injury, United States workers still suffer over 12,000 injuries per day (BLS,
The past decade has seen an increase in recognition and emphasis on promoting workplace
safety and increased understanding of the risks from exposure to occupational hazards. Each
year approximately 6 million U.S. workers suffer disease, disability, injury or death due to
exposure to such hazards. This averages to approximately 5 cases per 100 full-time employees
of which 93% are due to injury (BLS, 2003). Although government agencies such as the
National Institute for Occupational Safety and Health (NIOSH) and the Occupational Safety and
Health Admini station (OSHA) work on behalf of worker safety, complete elimination of risk is
difficult, if not impossible. Research has focused on limiting exposures, promoting safety
techniques, and work redesign.
Costs of Safety/Injury
The costs related to occupational hazards are enormous. Direct costs, which include
medical and hospitalization costs, as well as drugs and rehabilitative services, were
approximately $24 billion in 2000. Indirect costs are even greater. Indirect costs include loss of
wages, costs of fringe benefits, employer retraining and workplace disruptions and are estimated
to be over $100 billion per year (NSC, 2001). These numbers rival the cost of cardiovascular
disease and cancer, two of the greatest health issues facing Americans. In 2003, workers'
compensation picked up 27% of all costs, although taxpayers paid 18% of the total costs through
the Medicare, Medicaid and Social Security programs. Ultimately all Americans pay for the
costs of occupational injury and disease through lower wages, lower profits and higher costs for
consumers (Leigh, Markowitz, Fahs & Landrigan, 2000). Furthermore, these figures do not
include the cost burden on caretakers and relatives providing assistance to the injured and ill
(Arno, Levine & Memmott, 1999).
Health Care Workers
Health care workers are among those at highest occupational risk from injury or illness due
to the complexities of their work. Exposures include blood borne pathogens and biological
agents, chemical agents, physical stressors and psychosocial hazards. The health care industry
is one of the fastest growing occupational fields and because of its rapid growth the incidence
and prevalence of exposure has increased over the past ten years (NIOSH, 2005).
History of Occupational Health and Safety
The improvement in workplace safety along with the decline in occupational injuries and
fatalities over the last 100 years has been called one of the most important public health
successes of the 20th century (Centers for Disease Control [CDC], 1999). Work related injuries
were first identified by Bernardino Ramazzini in the late 17th century. Ramazzini is regarded as
the father of occupational medicine and wrote the first treatise on occupational health called
Diseases of Workers (Rosecrance & Cook, 1998). Ramazzini was the first to note the swollen
hands of bakers from repeatedly kneading dough, as well as the development of hunch back
posture in sedentary workers (Rosecrance & Cook, 1998). The Industrial Revolution of the
1800s ushered in factory work with its unsafe and unsanitary working conditions. Common law
in place at that time assumed some degree of worker risk was expected and protected
management from charges of negligence. However, a fire in a New York City factory in 191 1
that killed over 140 women and children brought public awareness and outrage to poor and
unsafe working conditions. The fallout from the Triangle Factory fire was the impetus for
reform in occupational safety (Cornell University, 2005).
The first survey of workforce health conditions took place in 1912 and focused primarily
on industrial accidents. Statistical record keeping, however, did not begin in earnest until the
1930s due to lack of conformity in regards to data (BLS, 2003). However, early surveys were
voluntary and included only fatalities or maj or injury. The passage of the Occupational Safety
and Health Act (OSH Act) of 1970 was a watershed event leading to improved recognition and
monitoring of workplace hazards. The OSH Act was designed to ensure safe and healthy
conditions for working men and women by establishing and authorizing enforcement of safety
standards in the workplace. The act also gave authority to states to enact programs of their own
and developed research and educational programs to increase knowledge of the subj ect as long as
they met federal standards (OSHA, 2005). The act broadly defined occupational hazards and
included not just death and injury but disease and disability as well. The act places the
responsibility for ensuring a safe working environment on management and includes criminal
fines for failure to comply with established standards. This emphasis was a shift from assigning
blame for worksite accidents on workers and instead focuses on indirect causes of accidents such
as unsafe conditions, poor match between worker and job and poor compliance with safety
standards. Yet despite these improvements in occupational safety, six thousand occupational
fatalities and 6 million injuries occurred in 2003 (BLS, 2003).
The Role of NIOSH and OSHA
The OSH Act of 1970 was the driving force behind the creation of both NIOSH and
OSHA. The National Institute for Occupational Safety and Health is a division of the Centers
for Disease Control and is responsible for conducting research in injury prevention, providing
training to health professionals, investigating occupational hazards, and developing educational
and occupational guidelines to protect workers. OSHA falls under the jurisdiction of the U. S.
Department of Labor and is the regulatory authority responsible for work site safety inspections,
development of safety standards and regulations, safety controls and worker education (CDC,
Of the 6 million injuries that occur each year, approximately one-third are disabling
injuries, meaning that the injury resulted in at least one day of absence from work. The other
two thirds of injuries are nondisabling and required no full days missed of work (Leigh et al.,
2000). Tenure and experience play significant roles in who suffers injury. New employees,
regardless of age, suffer a disproportionate number of injuries (Leigh et al., 2000). Men are
injured more frequently and more severely than women; Hispanics and blacks suffer more
injuries than non-Hispanic whites. Surprisingly, self-employed workers and those working in
small businesses are at high risk for suffering injury.
The most dangerous occupations in the United States are in the agriculture, mining and
construction industries, because of the large and heavy equipment involved. Types of hazards
vary depending upon the industry but are also influenced by economic structure, level of
industrialization, climate conditions and traditions of safety within the discipline. However,
recent data has shown the incidence of injury within these industries has declined over the past
decade. In contrast, the health care industry has noted an increase in reports of occupational
injury, in part due to its rapid growth (NIOSH, 2005). Nursing aides (NAs) are among workers
suffering the most disabling injuries. (Leigh et al., 2000).
Workplace hazards for health care workers are divided into four classifications: biologic,
chemical, physical and psychosocial. Biological agents include bloodborne pathogens such as
human immunodeficiency virus and hepatitis virus as well as airborne viruses such as
tuberculosis and severe acute respiratory syndrome (SARS). Needle stick injuries fall under
biological hazards. Chemical hazards include reaction to latex to harm from exposure to
antineoplastic drugs. Chemical hazards may also include gasses and smoke released during
surgical procedures (from anesthesia gases and cauterizations, respectively). Physical harm may
come from ionizing agents. Musculoskeletal injuries are also due to physical hazards, usually
from heavy lifting or pulling or repetitive motion. Violence in the workplace is another example
of physical harm. Lastly, stress at work, due to j ob responsibilities, excessive hours, and/or low
wages is considered a psychosocial hazard (NIOSH, 2001).
The most common workplace injury across all occupational groups that contributed to days
away from work in 2003 remained sprains and strains, particularly injuries involving the lower
back (BLS, 2005). When combined with bruises and contusions, fractures and cuts and
lacerations, sprains and strains accounted for nearly two thirds of causes for missed days.
Sprains and strains as the leading cause of missed days remained constant across all occupational
groups. Nursing personnel, particularly NAs, are among the workers having the highest
numbers of work-related injury due to overexertion (BLS, 2003).
Theories on Accidents and Safety
In order to prevent accidents and subsequently injuries, researchers have attempted to
define why accidents occur. Many theories have been postulated but none universally accepted.
Accidents are thought to be caused by a combination of several mistakes that occur during
different parts of the decision making process. There may be a long incubation period where the
same mistakes are made before an accident occurs. Accidents may be viewed as originating
from a technical error or human error perspective. Technical errors focus on design, engineering
and construction. Human error perspective focuses on intrinsic human factors such as cognition
and behavior. Accidents can be classified on a continuum of severity. The order of severity is
fatal accidents, serious accidents, first aid treatment, property damage and near accidents. Injury
implies some degree of damage to an individual. It has been estimated that less than 10% of
accidents result in injury (Taylor, Easter & Hegney, 2004).
One of the first theories on accidents came from Heinrich in 193 1. His domino theory
attributed 88% of accidents to unsafe acts of people, 10% due to unsafe conditions and 2% due to
act of God. He proposed a continuum of factors that cascade from one sequence to the next,
leading to an accident. His sequence begins with ancestry or social environment, followed by
worker fault, unsafe act combined with physical or mechanical hazard, accident, and finally
injury. Heinrich believed that removal any one sequence would stop the domino effect but that
eliminating the unsafe act was paramount to preventing injury (Taylor et al., 2004).
The multiple causation theory postulates that there are many contributory causes leading to
an accident and that it takes a certain combination of factors for an accident to occur. Multiple
causation theory categorizes causes into behavioral and environmental factors. Behavioral
factors include attitudes, skills and knowledge. Environmental factors include worksite hazards
and procedures that contribute to injury (Taylor et al., 2004).
The energy exchange model proposes that exchanged energy produces injury. The energy
arises from the type of hazard involved. The energy may be mechanical, electrical, thermal or
chemical. This theory is better suited for industrial accidents (Taylor et al., 2004).
The injury causation model uses a sequence of factors that contribute to an accident. The
sequence begins with an error that contributes to an accident. If a hazard is present, the sequence
may end with an injury. This model must include an error and a hazard for an injury to occur.
Usually the error leads to an unplanned event or accident. However, if a hazard is present, the
result may be an injury (Taylor et al., 2004).
Other theories of accident prediction include random chance theory that holds that
accidents occur simply by chance with no discernable pattern. Random chance theory suggests
that all workers are equally vulnerable to being involved in an accident and that preventative
interventions are without value. Biased liability theory holds that once a worker is involved in
an accident, he or she is more likely to be involved in another as compared to other workers
(Taylor et al., 2004).
Accident proneness has been defined to imply "that even when exposed to the same
conditions some people are more likely to have accidents than others, or that people differ
fundamentally in their innate propensity for accidents" (Shaw & Sichel, 1971 cited in Hansen,
1988). Accident proneness theory holds that there is a subset of workers who are more
susceptible to accidents.
The concept of accident proneness was introduced early in the 20th century, arising out of
concerns of the large number of accidental deaths and injuries occurring in British war
production industries (Haight, 2001). During this time injuries and accidents were accepted as
random events and their number could be predicting using mathematical principles established in
the Poisson distribution. However, investigation into industry accidents discovered that the
occurrence of accidents was not happening in normal distribution of Poisson' s formula.
Mathematicians began to develop a new model to predict accidents, which they labeled "accident
proneness." Thus began the search to measure and predict accident proneness. Early research
pointed to a correlation between accident proneness and "poor kinetic coordination and nervous
instability." However, attempts at finding a reliable measure over the next 30 years failed.
Accident proneness as a concept fell out of popular favor in the 1950s. Its reemergence in the
1970s corresponded with the drive to lower highway automotive accidents and make roadways
safer. Safer cars were built, roadways were improved, yet highway accidents remained. This
realization led the focus of interest being shifted to the driver as the cause of automobile
Marbe demonstrated in 1920 that people who had been involved in one accident were more
likely to be involved in another accident than someone who had been in no previous accidents.
Dunbar (quoted in Molnos, 1998) described accident-prone people as "impulsive, drawn to
adventure and excitement and always in search of immediate pleasure." The accident-prone
individual prefers spontaneity, resents authority figures, and is intolerant of discipline. He
proposed that these reactions were in response to a strict childhood upbringing (Molnos, 1998).
Molnos (1998) states that most accidents are subconsciously intended and are motivated by
guilt. The physical and psychological pain resulting from an accident is viewed as punishment to
the individual, thus relieving the guilt. However, she also hypothesizes that accidents are an
avoidance of responsibility.
Injury proneness, however, is more readily accepted as a personality trait but is recognized
as more difficult to amend. Other traits that are relative to an individual's injury proneness
include belongingness, self-efficacy, introversion/extroversion, perceptions of invulnerability,
conscientiousness, need for approval, impulsivity and emotional intelligence (Geller, 2004).
The Safe Environment
Development of a safe, injury free workplace requires focus on three aspects of interest:
the environment, the person and behavior. The environment includes the physical space,
instruments, tools and equipment as well as the safety climate within the setting. Behavior
includes the practices of all employees as each individual contributes to the overall safety of all
others. Lastly, attitudes, beliefs, and personality of employees play a critical role. All three
aspects interact and are dynamic and reciprocal (Geller & Wiegand, 2005).
What defines a personality is a subj ect as broad and diverse as the study of psychology.
Webster' s defines personality as "the complex of characteristics that distinguishes an individual;
the totality of the individual's behavioral and emotional characteristics" (Merriam-Webster
online, 2005). The word personality comes from the Latin word persona, meaning mask.
Therefore, personality can be viewed as the different masks a person wears. The term
personality can be used in two differing ways. The first refers to the distinct impression a person
makes on another. In this use, personality is similar to reputation and is defined from the
perspective of the observer. Secondly, personality may also refer to the essence inside each
person that explains behavior and creates an outward impression on others. In this sense,
personality is a person's identity that he or she holds within (Roberts & Hogan, 2001).
Personality has also been described as a "unique composite of inborn and acquired mental
abilities, temperaments attitudes and other individual differences in thoughts, feelings, and
actions" (Aiken, 1999). These individual characteristics are considered stable over time,
consistent and predictable (Aiken, 1999). The measurements of personality include affective
characteristics such as emotion and temperament, cognitive variables such as intelligence and
achievement as well as psychomotor skills (Aiken, 1999).
Personality assessments have been performed throughout time. There are references to
personality screenings in the Bible, in ancient Greek and Roman writings (the four humours) and
in manuscripts from the Middle Ages. Astrology, phrenology and palmistry are all attempts to
explain or predict human behavior. The first scientific theories of personality did not appear
until the late 19th and early 20th century and focused on intelligence and maladaptive personality
traits. The study of personality psychology evolved as a distinct discipline due to the work of
Allport in the 1930s. Although earlier psychological work included theories on personality,
character and abnormal psychology, Allport's text, Personality: A Psychological Interpretation,
united varying theories into a larger Hield of study. Allport viewed personality psychology as the
"study of the individual person" (Hogan, Johnson, & Briggs, 1997). Although controversies in
its definition and importance in mainline psychology remain today, advocates agree that
personality psychology features emphasis on three concepts: the whole person, motivation and
individual differences (Hogan, Johnson, & Briggs, 1997).
Biological and genetic theories on personality development
Proponents of biological and genetic theories argue that personality development follows a
predictable developmental process. The milestones may occur at slightly different rates and
times but they do occur in a predetermined order. Support for a behavioral genetics basis stems
from the high correlation of personality traits between monozygotic twins. Studies on
monozygotic twins have found, on average, a .50 correlation among personality traits. In
contrast, dyzygotic twins were found to have a .30 correlation among personality traits
(Matthews & Deary, 1998). Other studies on monozygotic and dyzygotic twins using the Five-
Factor Model (FFM) of personality have found evidence to support heritability for the
dimensions of conscientiousness, agreeableness, and openness to experience. Furthermore, the
evidence to support heritability for the dimensions of neuroticism and extraversion was
consistent and strong (Markon, Krueger, Bouchard & Gottesman, 2002). Studies on adopted
twins have found that on extraversion, children are more likely to resemble their biological
parents than adoptive parents (Matthews & Deary, 1998). Aggressiveness as a trait was found in
monozygotic twins to be due to heritable causes (58% of variance) (Beatty, Heisel, Hall, Levine
& LaFrance, 2002).
Evolutionary theories believe that certain traits or behaviors are found in individuals
because their presence was necessary for survival. Over generations, these traits continually
evolved and were reproduced. Aggression as a behavior was necessary to survive and adapt to
the environment. Aggressive animals assured the survival of their progeny, resulting in the
survival of strongest. Whether or not this explanation is applicable to human evolution is
uncertain (Hogan, Johnson & Briggs, 1997).
Eysenck attempted to discover a biological connection to personality development. He
argued that a person was either extraverted or introverted based on the stimulation level of his or
her reticular activation system (RAS) through cortical arousal. He hypothesized that introverts
had highly stimulated RAS, resulting in an avoidance of sensory stimulation, making the
individual more withdrawn and inward. Extraverts, by contrast, have an under aroused RAS and
as a result crave outside stimulation and excitement (Taub, 1998).
Hormones, testosterone specifically, has also been theorized to influence the development
of certain personality traits. Once again, aggression has been linked to high levels of
testosterone in both males and females. However, researchers are uncertain if aggressive
behavior is the result of increased levels of testosterone (Cohen-Bendahan, Buitelaar, van
Goozen, Orlebeke & Cohen-Kettenis, 2004).
Environmental theories on personality
Environmental theories support social, cognitive and cultural influences as having the
greatest impact on personality development. Their core beliefs hold that personality traits are
learned tendencies developed through modeling and reinforcement that occurs over a lifetime.
Behaviors are learned through experiences and by interaction with the environment (Nicholson,
Social learning theory emphasizes learning through social rewards, punishment, modeling
and reinforcement. Social cognition theory adds a cognitive component to learned behavior.
Personality development occurs when an individual performs an action in a certain situation.
The resulting feedback on that action modifies the behavior and influences its subsequent
repetition (Hogan, Johnson & Briggs, 1997).
Studies of monozygotic twins raised apart demonstrate the role of environment in
personality development. The correlations of monozygotic twins raised in separate environments
were twice as great as the correlations between similarly raised dyzygotic twins (Matthews &
Interactive effects of genetics and the environment are also felt to play a role in personality
development. A Swedish adoption twin study administered both a personality scale and an
environmental scale to identical twins raised apart. After controlling for main effects of genetics
and the environment, interactions were found to contribute 7% of the total variance in
personality scores (Matthews & Deary, 1998).
Culture has also been proposed as having considerable influence on personality
development. Cultural theories believe that personality dispositions are learned tendencies and
that the influences of the family and social network play the greatest role in their development
(Triandis & Suh, 2002).
Situation vs. person
The situation vs. person debate centers on how stable an individual's personality is over
time. Will an individual display the same personality traits consistently when put in similar
circumstances or will the circumstances dictate the behavior of the individual? Mischel first
proposed the situational view of personality traits in the 1960s. This undermined the accepted
view that personality dispositions were stable and could be used to explain behavior. His
proposal that situations dictated behavior was based on research that correlated personality and
behavior at .30. He argued that the small correlation meant that personality would not
consistently predict behavior and that behavior would vary based on situations. Other situational
researchers calculated that personality explained only 16-20% of behavior (Matthews & Deary,
1998; Nicholson, 1996).
Opponents to Mischel countered that correlations of .30-.40 actually predict an individual's
behavior 70% of the time. Furthermore, it was argued that personality was better at predicting
behavior over time and across all situations than in specific situations at specific times in part
because individuals generally chose the situation they are in (Matthews & Deary, 1998).
Development of traits
Can personality be changed? The five factor model developed by Costa and McCrae holds
that personality stems from biological causes (genes) that don't reach full maturity until
adulthood (1999, cited in Sirvastava et al., 2002). Hence, there should be little noted change in
adult personality over time. Costa and McCrae believe that personality traits are insulated from
the "direct effects of the environment" (1999, cited in Sirvastava et al., 2002) and reach maturity
around age 30 barring cognitive injury. Adult personality traits remain stable in adulthood
although they may change slightly in old age, again due to cognitive decline. Longitudinal
studies performed on both young adults and older adults have found personality relatively stable.
However, a literature review by Roberts and DelVecchio (2000, cited in Sirvastava et al., 2002)
came to a different conclusion. They found that scores for conscientiousness and agreeableness
tended to increase in adulthood while neuroticism scores decreased. Scores for openness to
experience demonstrated mixed results and scores for extraversion remained relatively stable.
Roberts argues that personality changes throughout life with different change patterns for men
and women. He notes that individuals are active participants in their life events and that those
experiences influences change. Work, marriage and parenthood are the biggest life changes that
affect adults. Because these events happen at different times for different people, the effect on
individual personality can occur at any age. These models do not presuppose consistency of
personality traits through genetics but instead believe consistency comes through the influences
of the social environment. Individuals are viewed as "open systems" that are both continuous
and changing at the same time This theory also proposes that societal, cultural and
psychological influences diminish with aging and maturity (Roberts, Walton & Viechtbauer,
Industrial/organizational (I/O) psychology is the study of the psychology of work and
includes the study of people, organizations, management, and behaviors (James & Mazerolle,
2002). Industrial psychology, also known as personnel psychology, includes the study of job
performance and analysis, staffing and recruitment, and abilities. Organizational psychology
studies organizational behavior and j ob satisfaction, leadership, motivation, j ob design, career
counseling as well as personality characteristics pertinent to a specific job (James & Mazerolle,
The application of personality psychology to industrial and occupational fields evolved in
the late 1980s. Employers traditionally used personal interview and cognitive testing to predict
worker performance and subsequent hiring. However, fears of discrimination led to the
development of instruments able to measure predicted performance based on personality
measures instead. The successful use of such an instrument on U.S. Army recruits broadened its
use in the general population. Researchers began using psychological measures to predict
outcomes such as job performance, job satisfaction and absenteeism of employees (Roberts &
The ongoing debate between nature vs. nurture hampered the emergence of personality
testing in the workplace. Mischel's conclusion that personality and behavior had only a .30
correlation became universally accepted in academia and led to the opinion that personality
could not be used to predict behavior. Furthermore, literature reviews performed in the 1950s
and 1960s used the same .30 correlation to claim that personality measures were of no value in
personnel selection (Matthews & Deary, 1998).
The 1970s were the nadir in the genetics vs. environment debate in regards to personality
measurement. Very few instruments were developed, funding was limited and morale low
among proponents of testing. The instruments most in use were good at identifying negative
traits such as neuroticism and hostility. However, if a person wasn't neurotic or hostile, the
instrument revealed very little about that person.
The development that most expedited the use of personality testing was the emergence of
personality assessment scales emphasizing normal personality. Prior to the 1950s, personality
assessment scales such as the Minnesota Multiphasic Personality Inventory judged abnormal
personality. In fact, the use of psychopathology measurements to predict performance of Office
of Strategic Services (OSS) agents during World Was II was widely viewed as unsuccessful The
first measurement developed to predict high level performance, the California Psychological
Inventory (CPI), was not developed until the 1960s (Matthews & Deary, 1998). Even today, the
stigma associated with personality assessment in the workplace traces back to the belief that
screeners are searching for abnormal or deviate traits.
Personality's reemergence as a desired tool for pre-employment screening came, not
surprisingly, from employers. Companies and businesses began emphasizing not only technical
skills and experience in job advertisements, but sought individuals with specific personality traits
as well. Among the desirable traits sought were initiative, integrity and self-discipline. The
stated goal of pre-employment screening was to predict the fit between the individual and the
occupation, thus sparing both the employer and the employee of an unsuitable relationship and
saving the employer money and the employee emotional distress.
Industrial and organizational (I/O) psychologists, who specialize in the study of human
behavior in the workplace, spurred the rediscovery of personality testing to predict j ob
performance. Four maj or factors contributed to the use of personality measurement in j ob
screening. The first involved cognitive testing. Intelligence is widely regarded as the greatest
predictor of occupational performance. However, cognitive measures have the potential to
discriminate against disadvantaged populations. Thus to alleviate the political and social
pressures on employers to avoid discrimination in the selection process, personality measures
became popular. Unlike cognitive measures, race and gender are neutral in personality measures
The second factor influencing the development of pre-employment screenings of
personality was the emergence of the Five Factor Model (FFM). The FFM consolidated
thousands of personality characteristics into the five broad dimensions of Extraversion,
Neuroticism, Agreeableness, Conscientiousness and Openness to Experience. The FFM resulted
in more consistent and reliable personality measures. The third influence was the publication of
the results of Proj ect A, a U. S. Army research proj ect conducted in the 1980s that used
personality testing for selection into entry level Army j obs. Lastly, two meta-analyses conducted
in the 1990s helped sway psychology opinion that personality measures provided a valuable aid
in explaining and predicting job performance (Goodstein & Lanyon, 1999).
Why use personality measures to prescreen employees? It is estimated that between 60-
90% of businesses use some form of pre-employment screening, often disguised as "employee
assessment testing" (Cox, 2003). Its purpose is to improve job performance and reduce turnover,
absenteeism, injury and poor customer service. Years ago employers looked only for outgoing,
skilled, motivated workers and discounted many others who didn't fit a particular profile. Now
businesses increasingly recognize that there is more than one type of ideal employee. Recruiting
now focuses on the employee-occupation fit. There are diligent and meticulous employees who
are excellent at accounting or computer programming but may have poor interpersonal skills.
Conversely, an extraverted employee may do well in customer relations or sales but lack
patience and attentiveness to detail needed for a desk j ob. High risk j obs, particularly in regards
to injury, may also be better suited to a specific type of person. Injury risk is related to j ob
performance in that compliance to established safety guidelines and careful diligence may
Ethical Considerations in Personality Testing
The two most important ethical considerations surrounding the use of pre-employment
screening instruments involve the issues of privacy and discrimination. On opposite sides of the
issues are workers who wish to be treated with dignity and with respect for their civil liberties
and employers who are under pressure to produce economically (International Labour Office,
A 1971 U.S. Supreme Court decision, Griggs vs. Duke Power, was an important landmark
in the use of pre-employment personality testing. Prior to this ruling, most personality testing
was done using measurements that had limited scientific validity and were used primarily as a
way to identify individuals with executive potential (Cox, 2003). Griggs vs. Duke Power
declared that use of personality measurements (as well as intelligence tests) were
unconstitutional and violated the 1964 Civil Right Act by "limit[ing], segregat[ing], or
classify[ing] employees to deprive them of employment opportunities or adversely to affect their
status because of race, color, religion, sex, or national origin" (U.S. Supreme Court, 1971).
However, the Griggs decision did allow professionally developed ability tests to be administered
if the tests were designed not to discriminate based on gender, race, religion or national origin.
Included in the decision was a prohibition on the use of polygraph testing to evaluate honesty
and reliability of employees. Yet, employers wanted to continue to evaluate these essential
characteristics in potential employees. The wish to identify individuals with certain desirable
traits led to the renewed interest in written personality measurements that assessed many of these
More recently, ethical concerns about pre-employment personality testing center on issues
of privacy. Do employers have the right to obtain pre-employment information, including
personality testing, from potential workers? If so, what kind of data protections must be set in
place to avoid violating confidentiality? Can the information be shared by human resources with
managers, directors and supervisors without discriminating against the worker? Can a worker
decline to be tested and still be competitive for employment? Will the obtained information be
used to eliminate applicants who pose potential liability risk? Is informed consent given by an
applicant in need of a job offered voluntarily and autonomously? Will the obtained information
be to the benefit or detriment of the applicant? These are questions that involve privacy,
confidentiality, fairness, informed consent or refusal and professional competence and
responsibility and have no clear cut answers (Levy, Wegman, Baron & Sokas, 2005).
Businesses do have a social and legal responsibility to provide for a safe working environment.
If a goal of pre-employment screening is to assure a good individual-job fit in hopes of
improving productivity and decreasing risk of injury, then employers may have a legal obligation
to pre-screen. Businesses have been held liable for injuries suffered in the workplace. If pre-
employment screenings have the potential to physically protect workers from injuries, then
businesses may have a legal and ethical obligation to perform them. Since the passage of the
OSH Act of 1971, the legal responsibility for providing workplace safety has fallen squarely on
the employer (Business for Social Responsibility, 2006; U. S. Department of Labor, 2000).
It has been proposed that pre-employment screening should be used only if it is an
appropriate preventive tool that addresses a specific workplace problem, the tests are known to
be accurate, reliable and have a high predictive value in the population screened and that medical
removal protection for earnings and j ob security is provided (Levy et al., 2005). In order to
protect workers, a proposed Bill of Rights for individuals who are subj ect to medical screenings
by their employers (current or potential) has been proposed. The Bill of Rights proposes that all
workers should have the right to 1) be told the purpose and scope of the examination, 2) be told
for whom the provider works, 3) be provided informed consent for all procedures, 4) be told how
results will be conveyed to the employer, 5) be told about confidentiality protection, 6) be told
how to obtain access to medical information in the worker's file and 7) be referred for medical
follow up, if necessary (Levy et al., 2005).
Purpose of the Study
The purpose of this study was to investigate the relationship between occupational injury
and personality characteristics. This study sought to identify an associative relationship between
personality dimensions of extraversion, neuroticism and agreeableness and occupational injury
reported by health care workers. Specifically, this study looked at sub traits of excitement
seeking, impulsiveness, compliance and angry hostility as targeted predictors of occupational
injury. This study compared the personality characteristics of health care workers who reported
injuries to the employer' s employee/occupational health office and to the investigator with
control subj ects who did not report injury (and were assumed to not have suffered an injury). If
such a predictive relationship can be established, it can be utilized to develop safety programs
geared toward those employees most at risk. The assumption of this study was that there was a
relationship between the personality characteristics and occupational injury. Therefore, the
research question for this proposal was as follows: Does possessing the personality traits of
excitement seeking, angry hostility, impulsivness and compliance increase the risk of a health
care provider suffering an occupational injury?
Significance of the Study
Despite the increase in the number of jobs in the health care sector over the past decade,
many specialties, particularly nursing, continue to suffer from acute staff shortages and an aging
population. Injuries sustained in the workplace frequently contribute to staff dissatisfaction and
lead to employee turnover and departures from the field altogether. Health care facilities such as
hospitals and nursing homes, recognizing the difficulty recruiting such staff, have begun to take
pains to retain employees by limiting exposure to injury. Furthermore, the costs, both direct and
indirect, resulting from worksite injuries, impact on facilities' ability to fund other needed
programs. The implementation of safety programs has eased the risk of injury somewhat.
However, neither behavioral nor engineering based programs have been successful in completely
limiting risk to employees. This study approached the issue from a person perspective. It
attempted to identify those employees most at risk for suffering injury by identifying specific
personality indicators. By doing so, it is hoped that interventions can be implemented tailored to
individuals most at risk.
Social Ecology Theory
Social ecology theory considers the interrelationship between personal and environmental
factors in human health and illness (Stokols, 1996). Its roots derive from public health and
epidemiology but now also encompass aspects of sociology, psychology and education (Green,
Richard & Potvin, 1995). Its use in injury prevention stems from the recognition that large scale
public health issues such as occupational injury are too complex to be explained by a single
orientation (Stokols, 1996). There is growing recognition that individual behavioral strategies to
encourage safe practice may be ineffective in a culture with an unsupportive environment or
unfavorable social norms (Schmid, Pratt & Howze, 1995). Therefore, interventions must be
directed at multiple levels and multiple sections (Green et al., 1995). These multiple levels range
from immediate peers and friends to cultural and organizational norms (McLeroy, Bibeau,
Steckler & Glanz, 1988). Multiple sections include home, work, community and national
environments (McLeroy et al., 1988).
Stokols (1996) describes several core principles of social ecology. The first principle
accepts that environmental settings have multiple physical, social and cultural "dimensions" that
influence health outcomes. By this principle the environment may have a cumulative effect on
health as well as a specific influence. A second principle of social ecology holds that personal
attributes such as genetics, psychological dispositions (personality) and behavior, along with
environmental factors, influence health. Therefore, environmental conditions that adversely
affect one individual may hold little significance to another. Consequently, researchers in social
ecology have found that compatibility with one' s surroundings is an important predictor of well-
being (Stokols, 1996). Social ecology also considers the premise of passive interventions in
addition to more traditional active interventions (Stokols, 1996). Active interventions require
that an individual perform voluntary and sustained effort to enact behavioral change. That is, the
individual must actively work to change behavior. However, behavioral interventions requiring
active participation have been difficult to sustain over prolonged periods of time. Passive
interventions, by contrast, can be more effective in that they target larger numbers of individuals
simultaneously and may not require voluntary or sustained effort on the part of the individual
(Stokols, 1995). Public service announcements promoting injury prevention are examples of
passive intervention. Lastly, social ecology approach to health promotion and injury prevention
is highly integrated with other disciplines. No one perspective is considered singularly.
Ecological approaches consider a variety of preventive strategies including public health and
epidemiology, behavioral and social sciences, and cultural change models.
The PRECEDE-PROCEED Model
This study was based on a model to for occupational injury prevention that considered the
influence of social, behavioral and environmental factors on health behavior. The model (Figure
1-1) is based on the PRECEDE-PROCEED process by Green and Kreuter (Green, 2004).
PRECEDE is an acronym for Predisposing, Reinforcing, and Enabling Constructs in
Educational/Environmental Diagnosis and Evaluation. PROCEED refers to Policy, Regulatory
and Organizational Constructs in Educational and Environmental Development and was added to
the framework to acknowledge the importance of environmental factors as determinates on
health and health behavior (Glanz, Rimer & Lewis, 2002). The framework includes two
important propositions. The first, that health and health risks are caused by multiple factors and
second, that because health and health risks are caused by multiple factors, interventions to affect
environmental, behavioral or social change must also be multi-dimensional and include
participation by the individual (Green, 1994). Because the risk of occupational injury continues
to remain high in health care workers, and individual behavioral interventions have demonstrated
limited sustained success, a social ecologic model was chosen to in an attempt to prioritize the
importance of injury prevention in the workplace.
A second theoretical framework for this study was modeled after the epidemiological triad
of host, agent and environment. The epidemiological triad views causation as being due to the
interaction of all three components. Under the right environmental conditions, a susceptible host
and an external agent may combine to cause injury (CDC, 1992). This study looked to find an
association between certain host factors, specifically personality traits, and injury. The external
agents will be the chemical, biological, physical or psychological hazards that health care
workers are exposed to daily. The environment includes both the physical and social
environment and includes job demands, staffing and shift work. The literature review for this
proposal was modeled on the epidemiological triad. A second proposed model for this study is
shown in Figure 1-3.
* H1: Health care workers who score high or very high in the sub facet of excitement
seeking (Extraversion) on the NEO Personality Inventory-Revised (NEO PI-R) personality
inventory will significantly and positively be at increased risk of suffering an occupational
* H2: Health care workers who score high or very high in the sub facets of impulsiveness
and angry hostility (Neuroticism) on the NEOPI-R personality inventory will significantly
and positively be at increased risk of occupational injury.
* H3: Health care workers who score low or very low in the sub facet of compliance
(Agreeableness) on the NEOPI-R personality inventory will significantly and positively
be at increased risk of occupational injury.
The predictive (independent variable) variables for this study will be the specific
personality scores obtained from each health care worker. The variables of interest are angry
hostility, impulsiveness, compliance and excitement seeking. The response (dependent) variable
will be injury or non injury.
Figure 1-1. Planning model to prevent occupational injury
Figure 1-2. The PROCEED-PRECEDE Model. Reprinted with permission from Dr. Lawrence
W. Green from http://1green.net/.
Figure 1-3. Epidemiological triad
REVIEW OF LITERATURE
The last 30 years have seen an increase in initiatives designed to improve safety and
decrease risk of injury in the workplace. Research conducted in this Hield has addressed the issue
from different perspectives. This literature review will investigate occupational injuries using
the epidemiological triad of host, agent and environment. The review will first focus on
biological, chemical, physical and social hazards that act as agents and contribute to worksite
injuries. Next the review will evaluate the environmental contributions to injury. These include
organizational structures and physical designs of the workplace that increase risk of exposure.
Lastly the review will evaluate injury risk due to host factors. Host factors include age, job
tenure and individual differences, specifically personality dispositions.
Traditionally in epidemiology, the term agent has referred to an infectious organism that
causes disease in the host. However, agent can also refer to chemical or physical exposures that
lead to increased susceptibility to injury (CDC, 1992). Health care workers face an array of
exposures ranging from biological pathogens such as hepatitis and tuberculosis, chemicals such
as latex and ethylene oxide, smoke released during surgical procedures and lastly from physical
violence that occurs at work (NIOSH, 2005). All of these exposures may lead to either injury or
Biological pathogens include viruses carried in blood and body fluids such as human
immunodeficiency virus and hepatitis virus, airborne pathogens such as tuberculosis and severe
acute respiratory syndrome (SARS). Transmission of bloodborne pathogens is generally caused
by splashes into mucous membranes or skin or through subcutaneous needle sticks. Medical
housestaff, nurse anesthetists, inpatient nurses, phlebotomists and surgical technicians are at
highest risk of transmission (Dement, Epling, Ostbye, Pompeii & Hunt, 2004). Waterson (2004)
and Sencan et al. (2004) both found that nurses had the greatest risk of sharps injury of health
care workers and those risk factors included giving shots in the patient room and in recapping
needles (Waterson, 2004). Physicians are at greatest risk for splash injury (Sencan et al., 2004).
The first hour of work and the last two hours of work provided the greatest exposure risk
(Marcias, Hafner, Brillman, & Tandberg, 1996). Having a permanent position and having less
nursing experience increased the risk of occupational blood exposure in French nurses (Rabaud
et al., 2000). Use of universal precautions, particularly wearing surgical masks, decreased the
risk of contracting SARS by health care workers (Wilder-Smith et al., 2005). Failure to use
appropriate respiratory protection when performing high risk procedures was also a risk factor
for acquiring tuberculosis infection in health care workers (Jelip et al., 2004). Overall, the
greatest risk of injury from exposure to chemical or biological agents is lack of safety training,
poor safety climate and poor safety practices in the organization (Gimeno, Felknor, Buau, &
Chemicals used frequently in the health care industry include ethylene oxide,
glutaraldehyde, anti-neoplastic drugs and latex. Ethylene oxide is a potential carcinogen used in
gas sterilizers. Glutaraldehyde is a cold sterilizer used to clean heat sensitive equipment and may
cause respiratory irritation. Drugs used to treat humans, particularly anti-neoplastic drugs, may
increase risk of genetic mutations and cancer in exposed workers. In addition, smoke emitted
during surgical/laser procedures contains chemicals harmful to human health. Lastly, latex is a
natural rubber used in health care equipment. Repeated exposure may lead to contact dermatitis
(NIOSH, 2005). Again, as was the case of biological agents, workers at greatest risk of injury
are those who fail to use appropriate safety practices.
Worksite violence is an increasing threat to the well being of health care workers.
Although there are four categories of violence (criminal intent, patient/staff, worker/worker and
personal), violent acts by patients and/or their caregivers is most common. Workers caring for
mentally ill or agitated patients or for patients suffering from dementia are at increased risk for
verbal or physical injury (McPhaul & Lipscomb, 2004). Other risk factors include caring for
patients with a history of alcohol/drug abuse, working in understaffed departments, especially
during visiting hours and meal times, working in overcrowded departments with long waits for
service, having inadequate security and unrestricted movement of the public, and having staff
poorly trained to deal with volatile patients (McPhaul & Lipscomb, 2004). Staffers having
increased patient contact suffered increased risk of injury whereas having strong supervisor
support decreased risk of suffering workplace violent injury (Findorff, McGovern, Wall,
Gerberich, & Alexander, 2004).
Stress is now widely viewed as an occupational agent that can contribute to injury
(Hemingway & Smith, 1999; Johnston, 1995; Salminen, Kivimai, Elovainio, & Vahtera, 2003;
Trimpop, Kirkcaldy, Athanasou, & Cooper, 2000). Stress over health care changes within the
work environment, such as inadequate staffing, contribute to increased musculoskeletal injury in
nurses (Lipscomb, Trinkoff, Brady, & Geiger-Brown, 2004).
Environmental factors contributing to accidents and injuries are extrinsic factors that
increase an individual's susceptibility and exposure risk (CDC, 1992). Environmental factors
may include the physical work surroundings as well as organizational structure of a business.
Regardless, environmental factors may increase the probability a worker has of suffering an
injury. The greatest risk factor for worksite injury is working in a hazardous job (Frone, 1998;
Kerr et al, 2001).
Much of the research on organizational factors contributing to employee injury has looked
at the effects of scheduling and staffing. In general, long working hours, extended working
hours and working overtime, contributes to increased risk of employee injury. Working
overtime hours has the greatest impact (61% increase), extended working hours (up to 12 hours
daily) increased risk of injury by 37% and working more than 60 hours per week increased the
risk by 23% (Dembe, Erickson, Delbos, & Banks, 2005; Trimpop, Kirkcaldy, Athanasou &
Cooper, 2000; Kirkcaldy, Trimpop & Cooper, 1997). Shift work is generally regarded as
working off hours (nighttime or evenings) or rotating shifts (alternating day and night shifts).
Shift work disturbs the body's normal circadian rhythms and can affect reaction time,
concentration and motivation, increasing risk of accidents and injuries (Canadian Centre for
Occupational Health and Safety, 1998). Night shift workers also report more fatigue, which has
shown to affect performance and safety (Muecke, 2004). Although comparison are difficult,
working off shifts does increase the risk of occupational injury and working rotating shifts has
the highest risk (Frank, 2000; Muecke, 2004). Carrying a high workload as well as working
under a poor organizational climate increased the risk of sustaining a needle stick injury in health
care workers (Clark, Rockett, Sloane & Aiken, 2002). Working full time as opposed to part time
also increased risk of injury (Engkvist et al., 2000). Institutions having lower workloads as
measured by staffing levels (high staffing equates to lower workload) also correlated to
decreased frequency of injuries (Cohen et al., 2004). Higher job satisfaction, higher control
over practice and lower j ob demands are associated with fewer on-the-j ob accidents and injuries
in nurses (Letvak, 2005).
High psychological job demands such as excessive work, conflicting demands and
insufficient time to complete tasks have been identified as risk factors for occupational injury
(Swaen, van Amelsvoort, Bultmann, Slangen & Kant, 2004). Swaen et al. (2004) also identified
low skill discretion (lack of challenging, creative work) and low decision authority (lack of
freedom, control to make decisions) as contributing to risk of injury. Both job satisfaction and
supervisor/coworker conflicts also were negatively associated with occupational injury (Kerr et
al., 2001; Swaen et al., 2004). Workers reporting a greater job satisfaction had a slight tendency
to return to work sooner after occupational injury (Murphy, 1994). An association between low
decision latitude, low skill discretion, problems in interpersonal relationships and occupational
injury in hospital workers has been observed (Salminen et al., 2003; Seago & Faucett, 1997).
However, they also found that highly monotonous work predicted injury. A study in the
Netherlands concurred, finding that high work pace and low intellectual discretion was
associated with increased musculoskeletal injury (Houtman, Bongers, Smulders & Kompier,
In health care facilities, the use of equipment to alleviate strain on individuals also appears
to decrease risk of injury. Lift teams and lifting devices shift the physical stressors away from
individual workers. Their use has been demonstrated to decrease the incidence and severity of
musculoskeletal injury in health care workers (Edlich, Winters, Hudson, Britt & Long, 2004; Li,
Wolf & Evanoff, 2004). Although lifting devices have been shown to decrease injury risk,
studies have shown them to be poorly utilized (Santaguida & Fernie, 1998). Use of lift teams
has lead to a decrease in the number of back complaints (OSHA, 2005). Facilities utilizing lift
teams reported reductions in lost time, restricted time, workers' compensation claims and injuries
to lifting team members. Electric and/or hydraulic beds have been utilized on nursing units with
some success. Electric beds allow the nurse to bring the patient to her/him rather than have the
nurse bend to reach the patient, thus alleviating strain on the back (Trinkoff, Brady & Nielsen,
2003). Workers receiving adequate training also were less likely to suffer injury (Engkvist et al.,
Other protective equipment utilized to reduce injury in health care workers includes special
eye, face and body wear designed to prevent splashes of biological or chemical agents into the
face or skin. Newer intravenous systems deliver fluids and medications to patients without use
of hypodermic needles and have resulted in fewer needle stick injuries and exposure to
bloodborne pathogens. Facilities using such systems reported less risk of needlestick injury to
employees (Wilburn & Eijkemans, 2004). Special sterilizers and scavenging systems have been
installed in hospitals to decrease exposure to chemicals hazardous to health (NIOSH, 2005).
Working in specific environments also increases risk of injury. Those departments
requiring staff personnel to perform heavy lifting demonstrate greatest risk of injury. Nurses
working on orthopedic, intensive care, neurology and surgery units had highest risk of suffering
back injuries whereas nurses who work in pediatrics and general medical units (which require
less heavy lifting) suffer fewer reports of injury (Goldman et al., 2000, Engkvist et al., 2000).
Having to perform patient transfers was among the highest predictors of suffering injury in
nursing personnel (Engkvist et al., 2000).
Host factors are intrinsic characteristics of individuals that may be due to genetics,
behavior or the environment. Host factors as causative agents vary based on the exposure,
susceptibility and response of the individual (CDC, 1992). Host factors include age, gender,
socioeconomic status, anatomy (height, weight), medical and psychological history and lifestyle
behaviors. Personality is an intrinsic characteristic that is defined as a host factor.
Research on age as a predictor of injury is mixed. Generally, younger workers are more
likely to be injured (Hansen, 1988). However, Kirschenbaum, Oigenblick and Goldberg (2000)
did not find age to be significant in testing a model of accident prediction. Harrell (1995) also
did not find age to predict injury among farm workers. However, a study of French laborers
found that increasing age decreased risk of injury but only until a certain age. After age 54, risk
and severity of injury increased, perhaps due to decreasing physical and cognitive ability
(Cellier, Eyrolle & Bertrand, 1995). Older women were also more likely to suffer serious injury
after slips, trips or falls at work (Cherry et al., 2005). However, increasing age was found to be a
protection against injury in nursing assistants (Meyers, Silverstein & Nelson, 2002). Whether
this is due to the healthy worker effect is unknown. The healthy worker effect is believed to
cause bias in studying age and longevity in workers. Workers who remain healthy, either
through genetics or healthy habits, are more likely to remain in the workforce. Therefore, it is
difficult to ascertain if increasing age is a protection against injury or if previously injured
workers have left the workforce at a young age (Arrighi & Hertz-Picciotto, 1994).
Gender as a predictor of injury also demonstrates mixed results. Males have found to be
injured more frequently but how much this is related to job risk is uncertain (Frone, 1998).
Iverson and Erwin (1997) controlled for j ob risk and found women to more likely be injured. In
contrast, Kirschenbaum and associates (2000) found male gender to be more predictive of injury.
Their study also controlled for job hazards.
Tenure has been researched extensively as a predictor of injury. Experience has generally
been inversely associated with accident involvement (Cellier et al, 1995, Liao et al., 2001).
However, in testing their predictive model, Kirschenbaum and associates (2000) did not find
tenure to be significant. Neither did Harrell (1995) in a study of farmers. A national
longitudinal study of American youth workers found that having longer tenure increased the
odds of being involved in an occupational injury or illness (Strong & Zimmerman, 2005). A
study of nurses found that less experienced nurses experienced greater frequency of injury (Arad
& Ryan, 1986).
Physical characteristics such as height and weight have been studied in relationship to
injury. Generally increasing body mass and weight is associated with greater risk of injury (Arad
& Ryan, 1986; Engkvist et al., 2000; Fransen et al., 2002). Life style choices such as poor diet,
lack of exercise and cigarette smoking may also be associated with increase risk of occupational
injury (Rosecrance & Cook, 1998; Sacks & Nelson, 1994). However, a study by Smith, Wei,
Kang and Wang (2004) found that occasional consumption of alcohol decreased the risk of
musculoskeletal injury. Income has also been noted to be inversely associated with development
of injury (Marras, 2000) although the model proposed by Kirschenbaum and associates (2000)
found higher paid workers more likely to be accident-prone. Previous history of injury is among
the highest predictors of injury (Marras, 2000; Venning, Walter & Stitt, 1987) as is poor physical
health (Frone, 1998).
Personality, Accidents and Injuries
The study of the influence of personality on injury or accident proneness has been
conducted in fields outside of health care. A variety of different scales have been utilized, each
with specific foci. The Minnesota Multiphasic Personality Inventory (MMPI) scale measures
deviant patterns of personality behavior while the Five Factor Model (FFM) personality scales
evaluate normal personality dispositions (Aiken, 1999). Some studies have focused on one
specific aspect of personality such as affectivity or locus of control. The results have varied
broadly but there have been some consistencies.
The MMPI consists of 10 clinical scales measuring psychiatric disorders and consists of
such titles as Hysteria (strong reaction to stress), Social Introversion and Psychopathic Deviant
(socially maladaptive) (Aiken, 1999). The MMPI is used primarily in the legal community but
has been utilized in occupational settings as well. Bigos (1991) found that the scales for
Hysteria, Psychopathic Deviate (anti-social behavior) and Schizophrenia, along with a history of
low back pain, predicted report of back injury in manufacturing employees. A separate report on
the same study looked specifically at scale 3 (Hysteria) and its subsets as predictors of back
injury and found that scoring high on the Lassitude/Malaise subscale provided the greatest
predictive value. The sub scales Denial of Social Anxiety and Need for Affection also predicted
report of back injury. However, the subscale Somatic Complaints failed to predict report of back
injury leading the authors to surmise that emotional and psychological factors play an important
role in predicting report of injury (Fordyce, Bigos, Battie & Fisher, 1992). Firefighters who
exhibited traits of social introversion (Social Introversion scale), maladjustment (Psychopathic
Deviate scale) and hysterical reaction to stress (Hysteria) were more likely to sustain an injury.
Furthermore, those firefighters who scored highly on social maladjustment (Psychopathic
Deviate scale) reported longer duration of injury. Surprisingly, firefighters who scored high on
the Schizophrenia scale reported shorter duration of injury (Liao, Arvey, Butler & Nutting,
2001). In summary, both the Hysteria and Psychopathic Deviate scales appear to predict
occupational injury. Scoring high on the Hysteria scale is associated with individuals who
demonstrate an excessive response to stress. They may appear well adjusted but under stress
may avoid responsibility. They may be socially isolated and immature (Fordyce et al., 1992).
High scores on the Psychopathic Deviate scale are associated with social maladjustment,
rebelliousness and impulsivity. It is hypothesized that individuals demonstrating these traits,
social maladjustment and social anxiety, may predict occupational injury and accidents by
engaging in reckless behavior.
The Five Factor Model (FFM) measures five broad dimensions of personality. The FFM
categorized thousands of existing personality measurements into five general "dimensions":
Extraversion, Agreeableness, Conscientiousness, Neuroticism and Openness (James &
Mazerolle, 2002; Roberts & Hogan, 2001). The FFM or "Big Five" was first proposed over 60
years ago but has only recently reappeared in the personality literature. Louis Thurston,
President of the American Psychological Association, in 1934 proposed narrowing down the
sixty adjectives used to define personality into five broad categories. His idea was not pursued
further until Catrell refined the broad categories in the 1940s. Finally in the 1980, Goldberg
aided by Costa and McCrae developed what is now widely accepted as the Big Five model
The dimensions are Openness to Experience, Conscientiousness, Extraversion,
Agreeableness and Neuroticism (James & Mazerolle, 2002). Extraversion includes facets of
assertiveness, activity, excitement seeking. High scores on extraversion are associated with
talkative and assertive individuals who are highly social. Agreeableness encompasses trust,
compliance, and altruism. In contrast, scoring low on agreeableness is related to poor
socialization and is associated with individuals who are angry, immature and impatient.
Conscientiousness values competence, self-discipline, and dutifulness. Self control, integrity and
honesty are important characteristics in this dimension. Neuroticism includes anxiety, hostility,
and impulsiveness. Persons testing high on neuroticism are viewed as fearful, anxious and
depressed and may engage in indecisive and withdrawal behavior. Lastly, openness to
experience consists of aesthetics, feelings, ideas and values (James & Mazerolle, 2002).
Individuals scoring high in openness to experience are more open to change, value differing
opinions and have a thirst for knowledge. Each dimension also consists of several sub facets that
measure more specific traits. As such the dimensions are very broad so there is great variation
among individuals, yet individual traits are stable over time. There has also been shown to be a
heritable component to the traits, meaning genes play a role in their development. The
dimensions have been tested and proven reliable within different cultures and languages
Although all dimensions have been studied in accident/injury research, conscientiousness,
extraversion, agreeableness and neuroticism have received the most attention. Employees who
are conscientious are expected to follow safety precautions and not act impulsively, lessening
risk of injury. They may also demonstrate greater self discipline and control. Extraverts are
viewed as more outgoing, social individuals. However, extreme extraversion, marked by over
confidence, intolerance and aggression, is thought to increase the risk of accidents due to risk
taking behaviors and carelessness. Introverts are more internally controlled and as such are
expected to be more vigilant in performing tasks (Hansen, 1988). Disagreeable individuals are
frequently viewed as social misfits and often exhibit negative, hostile emotions at work. Lastly,
individuals exhibiting neurotic behaviors such as anxiety and nervousness are thought to be more
susceptible to accidents.
These finding have been demonstrated in many studies although not all consistently within
the same study. An early study by Fine (1963) supported the hypothesis that extraverts would be
involved with more traffic accidents than introverts. Lajunen (2001) also looked at traffic and
occupational fatalities and found that extraversion positively correlated with traffic accidents
although neuroticism was negatively associated with accidents. Furthermore, he looked at
national traffic fatality data and found similar results. Extraversion again predicted traffic
fatalities within 34 countries. The results for neuroticism were mixed. He was unable to
demonstrate a relationship between personality factors and occupational fatalities. Extraverts
were also more likely to present for services at a minor trauma center seeking medical care
A cross sectional study of French nurses found that those scoring highest for disinhibition,
defined as "extroverted, socially disinhibited, stimulation-seeking behavior," and susceptibility
to boredom, were at risk for occupational blood exposure. Two other personality dispositions,
danger and adventure seeking and seeking new experiences, did not correlate with risk of
occupational blood exposure. Occupational blood exposure included both sustaining and
reporting exposure (Rabaud et al., 2000).
A study of university undergraduates found an inverse relationship between traits of
conscientiousness and agreeableness with "at fault" and "not at fault" accidents. Students who
demonstrated greater characteristics of conscientiousness and agreeableness were less likely to
be involved in any accident. However, no significant relationship between neuroticism,
extraversion and openness to experience with accidents was found (Cellar, Nelson, Yorke &
Bauer, 2001). Traits of extraversion (distraction), neuroticism (sensitization or high emotional
reaction to stressors) and avoidance coping styles predicted physical injuries necessitating
hospitalization in a general population (Marusic, Musek & Gudjonsson, 2001). The authors
hypothesized that individuals scoring high on extraversion were more easily distracted,
contributing to injury. They also hypothesize that individuals scoring high on sensitization
overestimate their tendency toward anxiety and thus are predisposed to misjudgments and
A model on worksite injury prediction found that individuals with greater social support
were less likely to be involved in accidents (Kirschenbaum et al., 2000), a finding confirmed in
other studies (Cellar, Yorke, Nelson & Carroll., 2004; Iverson & Erwin, 1997; Liao et al., 2001;
Marusic et al., 2001). When mediated by self-efficacy, agreeableness was associated with
involvement in fewer accidents, perhaps because of greater compliance with safety procedures
(Cellar et al., 2004).
The FFM has also been applied to children and has demonstrated similar results in
association with accident-related injuries. In a case control study of children ages 6-14 who
sustained accidental injury requiring overnight stay at hospitals, Vollrath et al. (2003) found that
cases scored higher on extraversion and lower on conscientiousness. This suggests that children
who are outgoing and energetic but possess lower concentration levels and lower desire to
achieve are more at risk for accidental injury. The same study found no association with
aggressiveness, impulsiveness or emotional instability to accidental injury.
The personality traits of managers also predicted injury and accident rates within their
departments. Twenty-three managers at a manufacturing plant were given a FFM personality
inventory. Managers demonstrating traits related to neuroticism (anxiety, nervousness) had
higher injury rates in their departments, perhaps because they push employees harder to increase
productivity which results in injury. In contrast, departments having managers who exhibit traits
of conscientiousness and extraversion were inversely associated with injury rates. Conscientious
managers are thought to be more likely to follow safety practices. Extraverted managers may
have greater communication skills with their employees, decreasing injury rates (Thoms &
Overall, research has supported the theory that extraverted persons are more vulnerable to
accidents and injuries. The evidence on neuroticism is less clear, however. Although increased
anxiety is felt to divert attention from tasks at hand leading to accidents, neurotic behavior may
instead lead to fewer accidents because heightened anxiety may contribute to greater
concentration (Hansen, 1988). Conscientiousness has consistently demonstrated an inverse
relationship with accidents. Conscientious individuals follow safety guidelines, are self-
disciplined and deliberate and are less likely to take risks contributing to accidents (Wiggins,
1996; Harrell, 1995). Agreeableness, specifically compliance, has been associated with
involvement in fewer accidents (Cellar et al., 2004).
Personality affectivity (mood) is an emotion based dimension through which a person
views his or her environment. Affectivity is based on experience and offers the individual a bias
through which life is perceived. An individual with a positive affect views life through positive
emotions and is seen as possessing an engaged, enthusiastic, confident personality. An
individual possessing a negative affect experiences negative emotions. He or she focuses on
disappointments and may exhibit feelings of anxiety and anger (Hogan, Johnson, & Briggs,
1997). Studies on affect and occupational injury report positive relationships. An early study by
Davids & Mahoney (1957) found that a negative attitude toward work increased susceptibility to
accidents. A positive affect, along with the ability to actively control the environment, decreased
the incidence of occupational injury among manufacturing workers in Australia. In contrast, a
negative affect, coupled with less direct coping mechanisms, increased the incidence of
occupational injury (Iverson & Erwin, 1997). Negative affect was also found to predict work
related injuries in adolescents although the results were not significant (Frone, 1998).
Possessing a negative affect is thought to be related to the sub facets of anger hostility, anxiety
and vulnerability of the neuroticism dimension (Garrity, 2003).
Locus of control (LOC) has been the focus of several studies looking at accident and
injury rates. The concept was first proposed by Rotter in 1966 and has been applied to many
behavioral fields. Locus of control measures an individual's expectancies of control.
Individuals possessing internal loci of control believe their own behavior is responsible for
rewards whereas a person with an external locus of control believes external forces outside their
control reward behavior (fate, chance) (Matthews & Deary, 1998). In theory, a worker who
feels little control over his work situation will more likely be involved in an accident.
Conversely, a worker who feels in control of his environment should take a more active role in
preventing accidents (Hansen, 1988). Weubker (1986) found that employees with severe
accident histories were more externally focused, although she did not control for demographics
such as j ob hazards, age and other factors. In a retrospective study of hospital employees,
workers with the highest external safety locus of control reported the most work injuries (Jones
& Wuebker, 1985). In a separate study workers reporting more control of their environments,
coupled with having a positive affect, had less risk of occupational injury (Iverson & Erwin,
1997). Other studies on both blue and white collar workers report similar findings (Janicak,
1996; Salminen & Klen, 1994). Thus, these studies support the theory that individuals possessing
an external locus of control have higher probability of being involved in an accident. The
concept of control is related to the sub facets of self-discipline and order found in the dimension
The concept of self-efficacy was studied along with the FFM in a survey of automobile
driving accident rates of college undergraduates. Self-efficacy is the belief an individual has in
themselves that they possess the skills and aptitude to accomplish certain tasks. The concept was
first proposed by Bandura in 1966 and has been applied to many behavioral settings. This study
sought to predict if students felt they could avoid experiencing a workplace accident in the next
ten years. Students scoring high on neuroticism along with low self-efficacy viewed themselves
as less likely to avoid being in an accident in the future.
Aggressive behavior is viewed as the tendency to act out anger and frustration (Hansen,
1988). Although numerous studies have linked aggression with automobile driving accidents,
aggression in association with occupational accident has not been heavily studied. Similar to
aggression is the concept of social maladjustment. This characteristic includes antisocial
behavior, immaturity, disregard for others, irresponsibility, hostility and authority problems
(Hansen, 1988). Liao et al. (2001) found that firefighters scoring high on social maladaptive
behavior reported longer duration of injury. An aggressive person can be viewed as socially
maladaptive and non compliant to society norms. Compliance is a sub facet of agreeableness.
Impulsiveness is a tendency to act quickly without thinking logically of possible consequences.
Impulsivity is a sub facet of the dimension of neuroticism and while neuroticism has
demonstrated mixed results in predicting accident involvement, a recent study did find an
association between impulsivity and accident involvement in nuclear industry workers (Garrity,
2003). Impulsivity has been studied extensively in pilots and drivers, but there is little research
linking impulsivity with occupational accidents and injuries. Impulsivity is a sub facet of
The Five Factor Model in the Workplace
The resurgence in the use of personality profiling in the workplace has occurred only in the
last two decades. Prior to this, many psychologists in the field of Industrial/Organizational (I/O)
Psychology viewed personality testing as not reliable or predictive enough for everyday use.
However, the refinement of the five-factor model of personality has encouraged its widespread
use for predicting success or failure in many occupational endeavors.
I/O psychologists acknowledge that competence in the workplace involves more than
cognitive intelligence. Human ability involves social and emotional intelligence as well (Roberts
& Hogan, 2001). However, success or failure in the workplace is also dependent upon multiple
factors beyond ability. The factors are work context, which includes interpersonal relationships,
physical work conditions and structural j ob characteristics: work style of the worker
(achievement oriented, conscientious, etc); and organizational context (type of industry, social
processes) (Roberts & Hogan, 2001). The importance of personality in work success or failure
is obvious. Employers want to hire individuals who will likely succeed in their role.
Furthermore, the advances in technology, including the increased complexities of tasks, requiring
longer and more costly orientations, make the desire for strong hires all the more desirable
(James & Mazerolle, 2002).
Beginning in the 1990s, studies have been undertaken evaluating the use of the FFM in
work related concepts such as job performance, job satisfaction and absenteeism.
Conscientiousness was shown to be a predictor of manager' s job proficiency (Barrick & Mount,
1991). Conscientiousness was also found to be significant, along with emotional stability, in
predicting job performance in supervisors (Ones, Viswesvaran, & Schmidt, 1993). In service
related fields, agreeableness and emotional stability (neuroticism) predicted performance
(McDaniel & Frei, 1994). A later meta-analysis also found conscientiousness had the highest
validity as a predictor of job satisfaction across occupational categories (Hurtz & Donovan,
2000). The FFM has also been applied to absenteeism. Judge, Martocchio and Thoresen (1997)
studied university employees and found that conscientiousness and extraversion predicted
absenteeism although the relationship was mediated by absenteeism history. Employees scoring
high on neuroticism did not predict absenteeism. However, high extraversion scores positively
predicted absenteeism and high conscientiousness scores negatively predicted absenteeism. A
meta-analysis evaluating the FFM and j ob satisfaction found the strongest predictors being
neuroticism and extraversion (Judge, Heller & Mount, 2002). Conscientiousness also correlated
with job satisfaction although not significantly.
The FFM has also been used to measure workplace counterproductive behaviors including
accidents, deviate behavior and turnover (Salgado, 2002). A meta-analysis found that
conscientiousness predicted deviate behaviors and turnover whereas extraversion, openness to
experience, agreeableness and emotional stability (neuroticism) predicted turnover with
emotional stability being the best predictor. There was no predictive value for accidents or
absenteeism in this study. Colbert et al. (2004) also found that employees testing low on
conscientiousness, emotional stability and agreeableness moderated the relationship between
negative perception of the workplace and workplace deviate behavior.
Summary of Literature Review
Although many of the findings of this literature review have mixed interpretations, several
variables stand out. Extraversion as a personality dimension consistently predicted an
individual's involvement in accidents. Whether this is due to an outgoing, active, excitement-
seeking, assertive nature or a lower level of vigilance is unclear. Few studies differentiated
between the different facets of extraversion. However, logic would dictate that individuals who
are less vigilant and careful in following safety behavior may suffer greater risk of accident.
Risk taking behavior, although an extreme of extraversion, is most likely the common trait
attributable to accident involvement.
The literature review on impulsiveness does appear to predict accident involvement,
although it has primarily been studied in drivers and pilots and not exclusively in occupational
settings. Individuals exhibiting impulsivity are more likely to react without careful thought or
deliberation. In the five factor model, impulsivity is a sub facet of neuroticism. Although
neuroticism as a personality dimension consistently fails to predict accident or injury,
impulsivity or recklessness is widely hypothesized to predict accidents.
Workers displaying traits of social maladaptive behavior including negative affect,
hostility, immaturity and irresponsibility are at risk for becoming injured on the j ob. These
finding were present on studies using both the FFM and the MMPI. Therefore, agreeableness
has been found to be negatively associated with accident involvement.
This study was conducted to evaluate the predictive nature of personality dispositions and
the occurrence of occupational injury, specifically in health care workers. It was conducted
prospectively and the data of interest was to be collected within a short time period of the injury
report (six weeks). By doing it so, the data was hoped to more accurately reflect subj ect
dispositions at the time of the injury and not be influenced by recall bias. Although difficulties
with recruitment necessitated expanding the timeframe from injury occurrence to study
participation to six months, it was still expected that this information would better predict the
occurrence of occupational injury.
This study was conducted using a case comparison design. Although there have been
prospective studies in this field, most research studying personality and worksite injuries have
used cross sectional designs with broadly generalized results. Most commonly the research has
collected both personality information and injury reports simultaneously and has depended upon
subject recall for injury history. Other studies obtained injury data retrospectively through
employer records. However, the personality inventory was not collected until much later and as
a result may not be as reflective of personality traits present at the time of injury. Lastly as
evident in the literature review, specific personality dispositions have not been studied as
frequently as the broad personality dimensions. It was hypothesized that by the use of a
prospective case comparison design, specific exposures could be tested against a comparison
group to more precisely define the significant personality characteristics contributing to worksite
accidents and injuries. Also, by obtaining the personality information near the time of the injury,
the identified traits will more likely reflect true dispositions.
The setting for this study was two community hospitals in the north central Florida region.
Ocala Regional Medical Center (ORMC) and Munroe Regional Medical Center (MRMC), both
located in Ocala, Florida, serve primarily Marion County. Both hospitals offer an array of
general services that include surgery, medicine, obstetrics and gynecology and pediatrics. ORMC
employs approximately 1000 staff that supports 270 hospital beds. Its occupational health office
reports approximately 100 injury reports per year. MRMC, the larger of the two facilities,
employs 2400 staff to support 421 patient beds and reports approximately 200 injury occurrences
annually. Each hospital has an occupational/employee health office/safety office that monitors
collection of data on accidents and injuries. Both hospitals employ a variety of health care
workers including nurses, nursing assistants, physical therapists, respiratory therapists, surgical
technicians, nurse practitioners and physicians.
Power analysis for this study using a level of significance of p<0.05 and a power of .80
required a sample size of 100 cases and 100 comparisons. However, due to recruitment
difficulties that will be discussed later, the study was halted with a final sample size of 29 cases
and 43 comparisons.
The inclusion criteria for cases were as follows:
* Health care workers who perform at least 80% of their working time in direct patient care
* Health care workers who work minimally 20 hours per week.
* Health care workers who report an injury claim to the occupational health office during the
study period of 6 months. (This was later modified and will be discussed in under
The inclusion criteria for comparisons were as follows:
* Health care workers who perform at least 80% of their working time in direct patient care.
* Health care workers who work minimally 20 hours per week.
* Health care workers who have not had an occupational injury claim within the previous six
months from the recruitment period.
The NEO Personality Inventory- Revised (NEO PI-R) was designed by Costa and McCrae
(Psychological Assessment Resources [PAR], Inc., 2005) to measure personality traits based on
the Five Factor Model of personality. The inventory was developed and refined over a number
of years beginning in the 1970s and today is widely viewed as a reliable, stable measure of
normal adult personality. The current revised version used in this study was first published in
1990. The NEO PI-R has been used in hundreds of studies and translated into over 25 different
languages. Its application includes Hields of clinical psychology, behavioral genetics and aging,
industrial/organizational psychology and health behavior. The measure consists of Hyve domains:
Neuroticism, Extraversion, Openness to Experience, Agreeableness, and Conscientiousness. In
addition, each domain consists of measures for six sub facets, providing specific information
about each domain. The scores on the six sub facets of each domain are tallied for an overall
domain score. The domain of Neuroticism consists of the sub facets of anxiety, angry hostility,
depression, self-consciousness, impulsiveness and vulnerability. Extraversion encompasses
warmth, gregariousness, assertiveness, activity, excitement-seeking and positive emotions.
Openness to Experience includes fantasy, aesthetics, feelings, actions, ideas and values. The
domain of Agreeableness carries the sub facets of trust, straightforwardness, altruism,
compliance, modesty and tender-mindedness. Lastly, Conscientiousness includes competence,
order, dutifulness, achievement striving, self-discipline and deliberation. The following are
examples of questions found in the NEO PI-R:
"I'm pretty set in my ways."
"I don't take civic duties like voting very seriously."
"I sometimes lose interest when people talk about very abstract, theoretical matters."
"In meetings, l usually let others do the talking."
"I have trouble resisting my cravings."
"I love the excitement of roller coasters."
"I'm hard headed and stubborn."
"I often get disgusted with people I have to work with."
There are 240 items (questions), as well as three validity items, and the inventory is
appropriate for adults with a minimum of a 6th grade reading level. Each item is scored on a five
point scale ranging from "strongly disagree" to "strongly agree." Once the inventory is
completed, the carbonated front sheet is removed, revealing a numbering system that divides the
questions by sub facet. The scores are then tallied and categorized into the appropriate sub facet
and domain. This provides a raw score for each domain and sub facet. The raw scores are then
transferred to a profile form for conversion from raw scores to t-scores. The t-scores of each sub
facet and domain are then listed into one of five levels: very high, high, average, low, and very
low. An individual completing the inventory can expect to receive both domain and sub facet
scores with an explanation that their scores are very high, high, average, low or very low
compared to the average person who has taken the inventory. Of all individuals who have taken
the NEO PI-R, approximately 38% score in the average range, 24% score in each of the high
range and low range and 7% score in each of the very high range or very low range (PAR, 2005).
The entire test can be completed in approximately 35-45 minutes. The internal consistency
coefficients for the domain scales range from 0.85-0.96. The internal consistency coefficients
for the facet scales range from 0.56-0.90 (PAR, 2005).
Job Relative Risk
Each subj ect was given a questionnaire to determine injury risk for their j ob. The
questionnaire consisted of eight questions that asked the subj ect to rate their risk on a 5-point
scale ranging from never to very often. The questions asked the subj ects to report their risk of
exposure to various biological, chemical and physical hazards occurring on the job. The
questionnaire, adapted from Frone (1998) and Garrity (2003), was labeled Job Risk
Questionnaire and is found in Appendix A.
Each subj ect completed a questionnaire seeking demographic information on age, gender
and tenure at the worksite. Information on number of hours worked weekly was also collected.
Age was collected in years and months and was self explanatory. Tenure was also collected in
years and months and was collected to represent number of years working in present position at
current hospital. This information was used to obtain frequency distributions on both the cases
and comparisons as well as to confirm eligibility criteria for the study (full time employees).
The subj ects also identified their j ob title, which was categorized by the investigator into one of
five groups: licensed nurse, unlicensed assistive personnel (UAP), technician, therapist and
provider. Because the array of job titles varies from each hospital, it was decided that the subj ect
would list their formal job title and then the investigator would determine which category was
appropriate. Licensed nurse include Registered Nurses and Licensed Practical Nurses.
Unlicensed assistive personnel included nursing assistants, patient care technicians and nursing
technicians. Surgical technicians, radiographic technicians (X-Ray technicians) and laboratory
technicians comprised the technician group. Therapists included respiratory therapists,
occupational therapists and physical therapists. Lastly, providers included physicians, nurse
practitioners, physician assistants and nurse anesthetists. This instrument was labeled
Demographic Questionnaire and is found in Appendix B.
Date collection procedures for this study changed and evolved over the course of subj ect
recruitment. Originally, cases were to be recruited through posted flyers in the occupational
health onfce waiting areas. The subj ect was expected to call the investigator and arrange for a
time and place to complete the instruments. The data collection was to be performed by the
investigator alone. However, this method proved unsuccessful and resulted in the recruitment of
only one subject. Therefore, it was decided to utilize the staff of the occupational health offices
to assist in recruiting subjects. The occupational health nurses would provide study information
to potential subjects when the employee presented with an injury. If the employee agreed to
participate, the occupational health nurses would give the subj ect a packet of the study
instruments to complete and return. Once the employee returned the completed study, they
would be given a $20 gift card as compensation. The investigator would then collect the
completed materials at a later date. Although this recruitment method eventually was put in
place, its implementation was delayed due to difficulty in obtaining Institutional Review Board
(IRB) approval for the change from the originally approved protocol. Proof of Health Insurance
Portability and Accountability Act (HIPAA) qualification of the occupational health nurses had
to be demonstrated to IRB prior to its granting approval. Because this process took several
weeks and multiple IRB revision submissions to facilitate, it presented a delay in case subj ect
recruitment. Therefore, it was elected to proceed with recruitment of comparisons. Controls
were to be recruited by posted flyers in break rooms of nursing units and departments. Although
this method was successful (due to the active involvement of the investigator in promoting and
discussing the study while posting the flyers), it also had the benefit of identifying potential
subj ect cases. Many employees (staff who qualified as either cases or comparisons) readily
consented to participate in this study once it was explained by the investigator. The investigator
was able to confirm if the subj ect met eligibility as a case or a comparison for the purposes of the
study. Upon agreeing to participate, each subj ect was then given a packet consisting of the three
instruments to complete. A time and date was arranged for the investigator to return to the
nursing unit, collect the materials, and provide the $20 gift card compensation. This method
gave the subj ect time to take the instruments home, complete them at their leisure and return
them when done. Although subj ect recruitment, particularly in regards to cases, did not occur
as originally planned, the three methods of case recruitment (posted flyer in the occupational
health office, utilization of occupational health nurses to recruit cases, and use of posted flyers in
nursing units and departments) eventually resulted in an adequate number of subj ects to
terminate recruitment after 6 months. Subj ect recruitment began August 2006 and ended January
Data Collection Procedures
Data was collected primarily by the investigator with the assistance of the occupational
health nurses. However, only the investigator tallied the instruments and recorded the results on
an Excel spreadsheet. Scoring the NEO PI R instrument required adding up points based on
subj ect response for the 240 items. Each question reflected one of 30 specific sub facets of the
instrument (anger hostility, compliance, impulsiveness, excitement seeking, for example).
Therefore tallying the instrument gave scores for each of the 30 sub facets. The six sub facets
for each particular domain (Neuroticism, Extraversion, Openness, Agreeableness, and
Conscientiousness) were then scored. The result was an overall domain raw score and raw
scores on each specific sub facet. Once the scoring of the instrument was complete, the raw
numbers were transferred to another instrument that provided t-scores for each domain and sub
facet. The final t-scores were then recorded on an Excel spreadsheet. The NEO PI R instrument
contains numerous reliability checks to ensure that the data was scored correctly. Scores were
added and then added a second time during the conversion of the raw data to t-scores.
The demographic information obtained from the Demographic Questionnaire (Appendix
B) was also tallied and coded. Age was coded by months. Tenure was also coded into months.
Job classification was categorized into licensed nurse, unlicensed assistive personnel,
technicians, therapists and medical provider. Gender was coded female or male.
The Job Risk Questionnaire (Appendix A) was not used in the final analysis of data. This
instrument was deleted at the advice of the statistician who advised that not enough subj ects
were recruited to allow for significant assessment of the tool. Therefore, the instrument was not
coded and analyzed.
Once the subj ect data was collected and tallied, the information was transferred to the
SPSS software program, version 11.5 (SPSS Inc., Chicago, IL) for analysis.
Procedures for the Protection of Human Subjects
This study was submitted to the Institutional Review Board (IRB) at the University of
Florida as an exempt study on April 7, 2006. After several resubmissions of requested
information, initial approval was granted on June 21, 2006. Changes in the recruitment flyer to
enhance subject recruitment necessitated additional revision submissions throughout the summer
and fall timeframes. Lastly, a revision requesting utilization of the occupational health nurses in
subj ect recruitment was submitted October 5, 2006. This revision received approval on
December 14, 2006 after the investigator provided proof of the occupational health nurses'
Once initial IRB approval was received the study was presented to the appropriate
personnel at the two participating hospitals. ORMC granted approval to use its facilities in
August 2006. This hospital does not have a formal nursing research committee in place.
Therefore the approval process required only going through the Chief Nursing Officer (CNO).
She did, however, refer the study to the hospital's legal department for approval. Once this
process was complete, the CNO notified nursing and departmental directors (including
occupational health) of the investigator' s approval to conduct the research. Therefore subject
recruitment began at ORMC in August 2006.
MRMC has a formal nursing research committee in place. A research protocol was
submitted and the investigator spoke to the committee in person twice about her proposed study.
Following nursing research committee approval, the study protocol was submitted to the various
hospital departments (including legal). Final hospital approval was granted August 2006.
Subj ect recruitment at MRMC began August 2006.
Informed consent was obtained from each subject participating in the study. However, as
an exempt study, a written consent form was not required. The subj ect' s participation in
completing the questionnaire conferred consent. Although this study required the collection of
confidential data, no intervention was conducted on human subjects. This study did not affect
any course of treatment, compensation or employment benefits of any participant. All data
collected was done so in a confidential manner. No identifying labels were placed on any of the
paperwork except when individual feedback on the personality inventory was requested. HIPAA
regulations were not violated as no information from the employer was given to the researcher.
The subj ects voluntarily contacted the researcher with their desire to participate in the study and
were allowed to withdraw at any time after consenting to the study if so desired. (Several
subj ects did just this. When the investigator returned to their nursing unit to collect the
completed materials, two subjects no longer wished to participate). No individual information
gathered during the course of the study was shared with employers. The final results of the study
will be provided to the participants and the employers if they desire to receive them.
Bioethical principles of self-determination, nonmaleficence, beneficence and justice were
observed during the course of this study. Self-determination or respect for autonomy recognizes
the dignity and autonomy of all individuals (Beauchamp & Childress, 2001). All subjects
enrolled in this study did so of their own accord. As the target population is a working
population, special populations such as children, the elderly and informed were excluded from
participation. Only English speaking and literate individuals were asked to participate in this
study; therefore, each subj ect was be able to consent to participate of their own will, without
requiring assistance from others.
Nonmaleficence is an obligation to protect research subj ects from harm due to their
participation in this study (Beauchamp & Childress, 2001). As no intervention was being
conducted as part of this study, no direct harm was possible. Indirect harm was possible,
however, due to political or economic pressures from the employer. However, care was taken to
respect both HIPAA regulations and subject confidentiality. All data was collected outside the
employment setting unless otherwise requested by the subj ect. The employer was not notified of
the subj ect' s consent to participate.
Beneficence in biomedical research requires that subj ects be awarded some benefit from
participating in this study (Beauchamp & Childress, 2001). As the purpose of this study was to
identify potential risk factors to occupational injury, it was to the benefit of subj ects to
participate. As health care workers, all subjects are at risk for suffering occupational injury in
the future. Findings from this study may lead to development of programs or policies that may
decrease risk of inj ury and ultimately benefit the subj ects.
The principle of justice requires that all subj ects be treated equally and that both the
benefits and burdens of the research be distributed fairly (Beauchamp & Childress, 2001). In
this study, all subj ects were given the same personality inventory and demographic
questionnaires and were offered the same monetary compensation. Therefore, the principle of
justice was not violated.
The first step in the data analysis was to obtain the univariate statistics (descriptive
statistics) on the subjects and measures. Means, standard deviation and percentages were
analyzed on the overall sample and the group (injured vs. non-injured). The same analysis was
conducted on the NEOPI-R for both the domains and sub facets. Bivariate analysis was
conducted between groups (injury vs. non-injury) by use of the t-test and cross tabs. The
analysis also looked at covariates such as age and tenure between groups. The t-test was chosen
because the variables of age and tenure were continuous. The bivariate analysis between groups
and job classification was evaluated by cross tabulation as job classification is a categorical
The Einal model was a regression model. The independent variables in this study were the
cumulative average t-scores on the sub facets of excitement seeking, impulsiveness, angry
hostility and compliance from the NEOPI-R. The dependent variable was injury vs. non injury.
Covariates such as age, tenure and job classification were computed in the analysis. The
domains of Neuroticism, Extraversion, Openness to Experience, Agreeableness and
Conscientiousness were also tested on the dependent variable. Because the dependent (or
response) variable is dichotomous, binary logistic regression was conducted on each group to
determine if there is a predictive relationship between the independent variables and injury
occurrence. Following the regression, each independent variable was analyzed for the
probability values for injury. No injury was used as the reference (constant). Odds ratios were
conducted on each predictive variable to determine probability of contributing to injury.
Power analysis was calculated by utilizing an effect size of 0.4. Although it is possible
that the effect size needed to be smaller based on previous studies that used an effect size of 0.25,
obtaining the required number of cases did prove to be difficult. Using an effect size of 0.4
required 100 subjects in each group.
ANALYSIS AND RESULTS
The data analysis for this study was conducted using the SPSS statistical software
program, version 11.5 (SPSS Inc., Chicago, IL). Descriptive statistics were first obtained to
provide summary measures for the data. Bivariate statistics were analyzed to compare the two
groups (injured vs. non-injured). Binary logistic regression was conducted to answer the
research question and test the study hypotheses.
As previously discussed, subj ects were recruited from two area hospitals in Ocala, FL,
Munroe Regional Medical Center (MRMC) and Ocala Regional Medical Center (ORMC). A
total of 80 subjects (34 cases and 46 comparisons) were recruited. Twenty nine cases (85%)
were recruited from MRMC and five (15%) from ORMC. Of the comparisons, 28 (61%) were
employed at MRMC and 18 (38%) at ORMC. For the overall sample, 57 subjects (71%) were
recruited from MRMC and 23 subjects (29%) from ORMC.
Of the 34 cases recruited, all but one was female. Of the five job classifications, all but
one case (a lab technician) was either a nurse or unlicensed assistive personnel (UAP).
Subsequently, to improve the statistical analysis, the one male case and one lab technician case
were eliminated from the data (which also resulted in the elimination of one male comparison
and two lab technician comparisons). Additionally, three case subjects were eliminated for
failure to complete questionnaires fully. The end result was 72 cases (29 cases and 43
comparisons) available for final analysis.
Seventy two subjects were included in the study. All (100%) were female. The mean age
and tenure for all subj ects is presented in Table 4-1. Of the 72 subj ects, 29 were in the injured
group and 43 subj ects were in the non-injured group. The breakdown between j ob classification
was approximately 66% nurse vs. 34% unlicensed assistive personnel in each group and the
overall sample. The mean age and tenure for each group (injured vs. non-injured) is also
presented in Table 4-1. The average age of the injured group was older than the non-injured
group although not significantly so. Additionally, the average tenure of the injured group was
greater than the non-injured group but also not significantly.
Bivariate statistics was performed between the two groups (injured vs. non-injured) to
evaluate differences between the two groups. Age, tenure and group were analyzed by use of the
independent samples t-test. Levene' s test for equality of variance was not significant for age
(p=0.552). Tenure was also not significant (p=0.969) for Levene's test. The t-test for equality of
means was also not significant for age (p=0.351) or tenure (p=0.471).
Cross tabs was conducted on job classification as it was not a continuous variable. Cross
tabulation between groups (injured vs. non-injured) and j ob classification (nurse vs. unlicensed
assistive personnel) was also not significant. Pearson's chi-square for this statistic showed a
significance of p=0.865.
Data analysis was also conducted on the personality domains of the NEOPI-R. Means and
standard deviations were obtained on the domains of Neuroticism, Conscientiousness,
Agreeableness, Openness to Experience and Extraversion for the overall sample as well as for
group (injured vs. non-injured). All means except for the Neuroticism score (44.93) for the non-
injured group fell within the average means for each personality domain. However, rounding up
this number places it within the average range. T-scores ranging from 45-55 are considered
average. This information is provided in Table 4.2.
The means and standard deviations were also obtained for the specific personality sub
facets within each personality domain. The means of the personality sub facets considered in
this study (excitement-seeking, impulsiveness, angry hostility and compliance) also scored
within the average means of the inventory for the overall sample and for each group (injured vs.
non-injured). T-scores in the 45-55 range are considered average. The t-scores for vulnerability
in the injured group (42.51) and trust in the non-injured group (42.86) were both in the low
range. The overall scores in the sub facets of vulnerability (44.69) and trust (44.95) also scored
in the low range. This information is found in Table 4-3.
The research question of this study asked: Does possessing the personality traits of
excitement seeking, angry hostility, impulsiveness and compliance increase the risk of a health
care provider suffering an occupational injury? For the purposes of this statistical analysis, the
research question was reframed as to ask if the sub facets of excitement seeking, angry hostility,
impulsiveness and compliance predicted group membership (injured vs. non-injured) after
controlling for age, tenure and job classification. Binary logistic regression is able to directly
predict the probability of an event occurring (Hair, Anderson, Tatham & Black, 1998). It is used
when the dependent variable has only two groups (the independent variables can be unlimited).
Logistic regression uses coefficients to predict the probability that an event will or will not occur.
It converts beta coefficients into logistic coefficients, also known as odds. Odds ratios calculate
the probability of an event occurring divided by the probability of no event occurring. A positive
coefficient increases the probability of an event occurring, whereas a negative coefficient
decreases the probability of an event occurring. The confidence interval (CI) provides an
estimated range of values. If the CI contains the value of 1.0 within its range, the variables
cannot be said to be a useful predictor (Hair et al., 1998).
To answer the research question, a binary logistic regression was conducted to assess if
excitement-seeking, impulsiveness, angry hostility and compliance correctly predicted subjects
as injured or non-injured after controlling for age, tenure and j ob classification (nurse vs.
unlicensed assistive personnel). As shown in Table 4-4, random assignment correctly predicted
59.7% of group membership (step 0). Age, tenure and j ob classification were next entered into
the first block of the regression equation and did not significantly predict group membership (chi
square = 1.17, p=0.76). Age, tenure and j ob classification correctly predicted 6.9% of the injured
subj ects and 93.0% of the non-injured subj ects, with an overall correct classification of 58.3%.
Excitement-seeking, impulsiveness, angry hostility and compliance were entered in the
second block of the regression equation and again did not significantly predict group
membership (chi square = 1.74, p= 0.78). Excitement-seeking, impulsiveness, angry hostility,
compliance, age, tenure and j ob classification corrected predicted 24. 1% of the injured subj ects
and 88.4% of the non-injured subj ects, with an overall correct classification of 62.5%.
Beta coefficients are coefficients resulting from standardized data and serve to calculate
the predicted change in the dependent variable. They may also be used to compare the relative
strength of the independent variables. The small standard errors indicate that the sample size
may have been too small to significantly predict group membership. The Wald statistic is used
with logistic regression to test for the significance of the coefficient. It is a test of individual
prediction. Table 4-5 demonstrates that none of the variables individually predicted group
membership (injured vs. non-injured). No variable had a significance ofp<0.05. Therefore, the
research question was not supported.
Table 4-1. Demographic statistics for job classification, age and tenure for the overall sample
and by group (injured vs. non-injured)
Overall Inj ured Non-Inj ured Significance
(N=72) (n=29) (n=43)
SD M SD
Table 4-2. Means and standard deviation of the personality domains for the overall sample and
by group (injured vs. non-injured)
46.19 10.55 47.04 11.50 44.93
51.87 8.62 51.41
48.51 10.29 47.81
50.36 10.68 48.86
52.52 10.56 52.06
Note: T-scores > 65
T-scores <35 :
10.96 53.20 10.09
very high. T-scores 35-44
very low. T-scores 45-55
low. T-scores 56-65 = high.
average. T-scores 35-44 = low
Table 4-3. Means and standard deviation for the personality sub facet scores for the overall
sample and by group (injured vs. non-injured)
Sub facet M SD
m Anxiety 48.66 10.67
Angry Hostility 48.16 10.03
Depression 47.01 10.05
Self-Consciousness 46.76 9.90
Impulsiveness 47.19 10.27
Vulnerability 44.69 8.90
very low. T-scores
Note: T-scores > 65= very high. T-scores
56-65=high. T-scores 35-45=low.
Table 4-4. Classifieation by group (injured vs. non-injured) by excitement-seeking,
impulsiveness, angry hostility and compliance (Step 2) after controlling for job
classification, age and tenure (Step 1)
Observed Group Predicted Group
Inj ured Non-injured % Correct
Step 0 Injured 0 29 0.0
Non-injured 0 43 100.0
Step 1 Injured 2 27 6.9
Non-injured 3 40 93.0
Step 2 Injured 7 22 24.1
Non-injured 5 38 88.4
Table 4-5. Logistic regression on excitement seeking, impulsiveness, compliance and angry
hostility predicting group membership (injured vs. non-injured) after controlling for
age, tenure and job classification (nurse vs. UAP)
95% CI for
Step Predictors B S.E. Wald Sig. Exp(B) Lower Upper
Block 1 Age -.002 0.002 0.577 0.447 1.00 0.99 1.00
Tenure -.001 0.004 0.045 0.833 1.00 0.99 1.01
Job Classifieation -.252 0.541 0.217 0.641 0.77 0.27 2.22
Block 2 Age -.002 0.002 0.727 0.394 1.00 0.99 1.00
Tenure -.001 0.004 0.041 0.839 1.00 0.99 1.01
Job Classifieation -.353 0.602 0.344 0.558 0.7 0.12 2.28
Impulsiveness -.022 0.032 0.463 0.496 0.98 0.91 1.04
Angry Hostility -.018 0.040 0.056 0.813 1.01 0.93 1.10
Compliance -.018 0.030 0.342 0.342 0.98 0.93 1.00
Excitement seek. -.029 0.032 0.833 0.833 0.97 0.93 1.04
CONCLUSIONS AND RECOMMENDATIONS
Occupational injuries to nursing personnel and other medical providers will continue to
take a significant toll on the health care industry over the next decade. Nursing shortages,
whether due to burnout, retirement or injury, continue to plague both hospitals and nursing
homes. Loss of experienced personnel from injury or illness impacts the provision of quality
patient care and ultimately leads to dissatisfied consumers and staff and higher operating costs.
The issue of staff safety has not gone unnoticed. Numerous safety programs have been
implemented and although these programs have improved worker safety, risk has not been
eliminated fully. Therefore, this study looked to address the issue of worker safety from the host
or person perspective. It hoped to identify staff at higher risk for occupational injury by testing
specific personality traits and ultimately lead to the development of tailored safety programs
aimed specifically at the individual.
This study used a case comparison design to evaluate if possessing the personality traits of
excitement seeking, impulsiveness, compliance and angry hostility increased an individual's risk
for sustaining an occupational injury. Seventy two health care workers met the study criteria.
Discussion of Findings
The 72 subj ects were divided into either an injury or non-injury group. This was done
based on the subj ect' s report of an occupational injury within the preceding six months.
Although there were more subj ects in the non-injury group (43 vs. 29), the bivariate statistics did
not demonstrate significance between the two groups. The division between nurse vs. unlicensed
assistive personnel in each group was similar (65.5% nurses vs. 34.5% unlicensed assistive
personnel for the injured group) vs. (67.4% nurses vs. 32.6% unlicensed assistive personnel for
the non-injured group). The injured group was slightly older than the non-injured group (42.32
years vs. 39.89 years). Although the difference was not significant, this finding is in contrast
with the literature that found older workers less likely to suffer occupational injury (Hansen,
1988). Meyers, Silverstein and Nelson (2002) also found increasing age protected nursing
assistants from injury. The injured group also had a longer tenure (8.23 years) vs. the non-
injured group (7.07 years). Again, this result was not significant but was inconsistent with the
literature review that found less experienced nurses more likely to suffer occupational injury
(Arad & Ryan, 1986).
The analysis of the NEOPI-R revealed consistently average scores on all domains and sub
facets. T-scores on the Hyve domains of Neuroticism, Extraversion, Openness to Experience,
Agreeableness and Conscientiousness were found to be within the average range on the five
point scale (very high, high, average, low, very low) for both the overall sample and each group.
With the exception of the sub facets of vulnerability and trust, all other sub facet t-scores were
found to be within the average range for the overall sample and each group. The injured group
scored low for vulnerability as did the overall group. The non-injured group scored in the
average range. This finding is not surprising given that many individuals who have been injured
Eind themselves in fear of re-injuring themselves (Dembe, 2001; Tarasuk, & Eakin, 1994).
Furthermore, having to deal with the medical and occupational staff may leave the injured
employee feeling helpless and not in control. These Eindings may leave the member with
feelings of vulnerability. The non-injured group (as well as the overall group) scored low on the
trust sub facet. The injured group scored in the average range. This is perhaps a reflection on a
non-injured individual's completing the questionnaire and having concerns regarding what the
information may be used for (Das & Teng, 2004).
Binary logistic regression was used to predict membership into the two groups (injured vs.
non-injured). Group alone predicted 59.7% of group membership meaning that chance alone
would predict nearly 60% of group membership. However, controlling for job classification, age
and tenure did not improve prediction into groups (p=0.70). Controlling for these variables only
correctly predicted 2 of 29 injured subj ects (6.9%) and 40 of 43 (93%) of non-injured subj ects;
the overall predictive percentage remained at 58.3%. However, group prediction was improved
by entering impulsiveness, compliance, angry hostility and excitement seeking into the model.
These variables improved the prediction of injured cases from 2 of 29 to 7 of 29 (an increase to
24%). However, prediction of non-injured subjects decreased to 88.4% (38 of 43 subjects). The
overall prediction did, however, improve to 62.5%. Although this result was not significant
(p=0.78), the model was able to improve group prediction from 6.9% to 24.1%.
Reviewed individually, none of the personality variables (compliance, angry hostility,
excitement seeking and impulsiveness) predicted group membership. The significance ranged
from p=0.342 (compliance) to p=0.833 (excitement seeking). Because none of the variables
significantly predicted group membership, odds ratio could not be conducted on the findings.
None of the hypotheses presented in this study were supported. The following is a
discussion of the individual hypotheses:
*H1: Health care workers who score high or very high in the sub facet of excitement
seeking(Extraversion) on the NEOPI-R personality inventory will significantly and
positively be at increased risk of suffering an occupational injury.
This hypothesis was not supported by the results of the study. The sub facet of excitement
seeking did not predict group membership (p=0.83). However, as will be discussed later, it is
possible that low observed power lead to a Type II error.
* H2: Health care workers who score high or very high in the sub facets of impulsiveness
and hostility (Neuroticism) on the NEOPI-R personality inventory will significantly and
positively be at increased risk of occupational injury.
This hypothesis was not supported by the results of the study. The sub facets of
impulsiveness (p=0.430) and angry hostility (p=0.813) did not significantly predict group
membership. Again, the possibility of a Type II error may have influenced results.
* H3: Health care workers who score low or very low in the sub facet of compliance
(Agreeableness) on the NEOPI-R personality inventory will significantly and positively be
at increased risk of occupational injury
This hypothesis was not supported by the result of the study. The sub facet of compliance
(p=0.342) did not significantly predict group membership. Again, a low sample size may have
contributed to a Type II error.
Limitations of Study
Study Design Limitations
All studies have some degree of limitation. One limitation of this study included the
difficulty in determining when an actual injury occurred. Many injuries are cumulative injuries
and their report by employees is influenced by a variety of factors. It is difficult to pinpoint
exactly when an injury is sustained. Therefore, this study accepted that time of sustaining of an
injury and time of reporting an injury may differ. For the purposes of this study, report of injury
was found to be more easily quantified. However, even this became difficult to determine. The
initial plan for subj ect recruitment was to solicit subj ects through the occupational health office
by posted flyer. The expectation was that an injured employee would see the flyer and contact
the investigator to participate. Because this method proved unsuccessful, utilization of the
occupational health staff ensued but even this did not provide a large number of subj ects.
Ultimately it took the active participation of the investigator to recruit an adequate number of
subj ects as well as extending the timeframe from injury report to participation in the study from
six weeks to six months. Therefore, the final personality scores are not as reflective of
personality at the actual time of injury as hoped.
As mentioned, subject recruitment for cases proved to be extremely difficult in this study.
The final number of cases (29) fell significantly short of power analysis projections (100).
Fortunately occupational injuries are not a routine occurrence. Estimates for the number of
expected injures for the two participating hospitals was 300 per year, perhaps only half of these
being health care workers. Yet in six months of subj ect recruitment, a far smaller number was
recruited. As with all studies, the more involved the investigator can be with subject recruitment
the more successful recruitment will be. Unfortunately HIPAA regulations limited the
involvement of the investigator until the subj ects contacted her.
The decision to remove the male case subj ect and the lab technician case subj ect served to
improve statistical analysis. However, it limited the study's results applicable to only women
and either nurses or unlicensed assistive personnel. Although the goal of this research was to test
the association between personality traits and occupational injuries in all health care workers, the
final study was limited to only women and nursing personnel.
The investigator also used a convenience, nonrandom sampling technique for recruiting
subj ects. Most of the subj ects were recruited directly by the investigator during her presence on
the nursing unit. Frequently the investigator would arrive on a nursing unit to collect a
completed questionnaire packet only to find other staff wanting to participate. This often
resulted in more subjects being recruited from some nursing units than others. Furthermore,
although the investigator attempted to recruit subj ects from all shifts, inevitably most of the
subj ects worked on the day shift. Use of this type of sampling can threaten the external validity
of the study.
There is also a great deal of literature addressing psychosocial aspects relating to the work
environment. Work variables such as role ambiguity, job boredom, autonomy, job
dissatisfaction and coworker/supervisor support have been identified as influencing j ob
performance, including susceptibility to accidents (Barrick & Mount, 1991; Judge, Heller &
Mount, 2002; Judge, Martocchio& Thoresen, 1997). This study did not control for these
Statistical Analysis Limitation
As discussed previously, power analysis for this study, based on the number of variables
under consideration, recommended a sample of 100 cases and 100 controls. The final subject
number was 29 cases and 43 controls. It is possible that the significance in this study was not
found due to inadequate sample size. Type II errors occur when study findings fail to reject a
false null hypothesis. This mean that a null hypothesis is not true, but the study fails to support
this. For this study, the null hypothesis would state that there is no association between the
personality traits of compliance, excitement seeking, angry hostility and impulsiveness and
injury occurrence. A Type II error would rej ect this hypothesis even if the null hypothesis was
false. Type II errors generally occur from not having enough subjects in the sample size to
sufficiently test the hypothesis.
Strengths of the Study
One of the strengths of this study included the consistency within the target population of
the two participating hospitals. Both hospitals provide comparable services and employ similar
staff. This provided a study sample that was relatively consistent in tenure, age and gender.
Furthermore, the instrument in use, the NEOPI-R, is a well established and easy to use
personality inventory that has strong reliability and consistency coefficients. The scoring of the
instrument answer sheets was performed by the investigator alone, guaranteeing consistency in
Clear implications for clinical practice cannot be drawn from this study. Based on this
study, there was no association between specific personality traits and risk for occupational
injury in health care workers. However, nurses and other health care workers do continue to
receive injuries on the job, and many studies have demonstrated an association between
personality traits and injuries. Although the NEO PI-R is a respected and commonly utilized
personality inventory, there was no documentation in the literature of its use with health care
workers. In fact, most of its use in Industrial/Organizati onal Psychology has been in identifying
leadership potential, person-job fit, job performance, job absenteeism, etc. There is only one
study that used the NEO PI-R to predict occupational injury (Cellar, Nelson, Yorke & Bauer,
2001), and the subj ect focus of that study was undergraduate students. The maj ority of research
that studied occupational injury used one or two small personality measures (similar to sub
facets) to predict occupational injury rather than a large, comprehensive personality inventory
(Frone, 1998; Iverson & Erwin, 1997; Janicak, 1996; Marusic, Musek & Gudjonsson, 2001).
Additionally, none of these researchers studied health care workers. This investigator found only
one study that looked at the association between personality and injury in health care workers
(Rabaud et al., 2000), and this study did not utilize the NEO PI-R in evaluating the nurses'
It is possible to conclude that the research findings are a result of a poor choice in the
utilization of the personality inventory. The NEO PI-R has not been widely utilized in
predicting occupational injury. Different study results may have been found by use of different
personality instruments, specifically those focusing on the sub facets in question. Furthermore,
unlike professions such as the military, NASA and nuclear energy, health care workers have not
been subjected to personality study. It is possible that because health care workers come from a
variety of cultures, educational levels, socioeconomic groups, etc., there is no common thread
among them that can be used to predict injury occurrence.
Recommendations for Further Research
Unfortunately this study did not find significance in its hypotheses that possessing the
personality traits of excitement seeking, compliance, impulsiveness and angry hostility predict
occupational injury in health care workers. However, as the literature review pointed out, there
have been numerous studies demonstrating the association between personality traits and
occupational injury in other fields. The use of the NEO PI-R inventory in this particular study
has also been discussed including the possibility that it was the wrong inventory to use in this
subj ect group and for this outcome. However, because the prevention of occupational injury is
such an important goal, further research on health care workers is imperative. If possible, this
study may be continued or replicated in the future, particularly with no time constraints and
perhaps utilizing a larger number of hospitals. It would be particularly helpful to have
occupational health staff as co-investigators of the study to enhance subj ect recruitment.
Additionally, it may be beneficial to identify the personality characteristics of health care
workers in general before beginning any study on occupational injury. There may be specific
traits that are common among health care workers or within their specific subset (nurses,
physical therapists, etc.). This background information can then be used to study the association
between these characteristics and occupational injury and hopefully discover positive findings.
Lastly, a different instrument, particularly one that studies specific indicators or traits, may
provide the best tool for conducting this research.
Implications for Clinical Practice
Because this study did not Eind significance in any findings, its application to clinical
practice is difficult to ascertain. However, there has not been an extensive amount of research
studying personality traits of health care workers and their association with occupational injury.
Most similar studies were conducted in Hields such as agriculture and construction that utilize
heavy equipment. Personality screening is conducted routinely in the military and other
professions requiring extensive training. If the cost of occupational injury continue to plague the
health care industry, it would not be unreasonable to seek valid screening tests assessing an
employee's risk of occupational injury before assigning individuals to work in high risk areas.
JOB RISK QUESTIONNAIRE
Please answer the following questions to the best of your ability. The information is confidential
and will not be shared with anyone at your job. You are also asked not to share the information.
Please circle the response that best describes how often in the last six (6) months you have
experienced each situation in your present j ob.
1. Have you been exposed to situations on the j ob that could cause you to lose your balance
1 2 3 4 5
never seldom sometimes often very often
2. Have you been exposed to situations on the j ob that expect you to carry, lift, push or pull
1 2 3 4 5
never seldom sometimes often very often
3. Have you been exposed to equipment and machinery on the j ob that could injure you?
1 2 3 4 5
never seldom sometimes often very often
4. Have you been exposed to needles and other sharp obj ects on the j ob that could injure
1 2 3 4 5
never seldom sometimes often very often
5. Have you been exposed to blood and body fluids on the job that could injure you?
1 2 3 4 5
never seldom sometimes often very often
6. Have you been exposed to airborne agents on the job that could injure you?
1 2 3 4 5
never seldom sometimes often very often
7. Have you been exposed to chemicals on the job that could injure you?
1 2 3 4 5
never seldom sometimes often very often
8. Have you been exposed to violence on the j ob that could injure you?
1 2 3 4 5
never seldom sometimes often very often
Please provide the following information to the best of your knowledge. The information is
confidential and will not be shared with anyone at your j ob. You are also asked not to share the
1. Please circle your gender: Male Female
2. Please list your age in years and months: years months
3. Please list number of years working in present position: years months
4. Number of hours worked weekly (average):
5. Have you suffered an occupational injury in the past six (6) months? Yes No
LIST OF REFERENCES
Aiken, L.R. (1999). Personality assessment: M~ethods and practices (3rd ed.). Seattle, WA:
Hogrefe & Huber Publishers.
American Nurses Association. (1997). Position statements: Registered nurse utilization of
unlicensed assistive personnel. Retrieved November 26, 2005 from
http://www.nursingworld. org/readroom/position/uap/uapuse.htm Silver Springs, MD:
Arad, D. & Ryan, M.D. (1986). The incidence and prevalence in nurses of low back pain. The
Australian Nurses Journal. 16 (1), 44-48.
Arno, P.S., Levine, C. & Memmott, M.M. (1999). The economic value of informal caregiving.
Health Affairs, 18(2), 182-188.
Arrighi, H.M. & Hertz-Picciotto, L. (1994). The evolving concept of the healthy worker survivor
effect. Epidemiology, 5 (2), 189-196.
Barrick, M.R. & Mount, M.K. (1991). The big-five personality dimensions in job performance: a
meta-analysis. Personnel Psychology, 44, 1-26.
Beatty, M.J., Heisel, A.D., Hall, A.E., Levine, T.R. & LaFrance, B.H. (2002). What we can learn
from the study of twins about genetic and environmental influences on interpersonal
affiliation, aggression and social anxiety? A meta analytical study. Conanunication
M~onoguraphs, 69(1), 1-18.
Beauchamp, T.L. & Childress, J.F. (2001). Principles of bionzedical edrliL (5th ed.). New York
City: Oxford University Press.
Bigos, S.J., Battie, M.C., Spengler, D.M., Fisher, L.D., Fordyce, W.E., Hansson, T.H.,
Nachemson, A.L. & Wortley, M.D. (1991). A prospective study of work perceptions and
psychosocial factors affecting the report of back injury. Spine, 16(1), 1-6.
Bureau of Labor Statistics. (2005). Sprains and strains most conanon workplce injury.
Retrieved July 23, 2005 from http ://www.bls.gov/opub/ted/2005/mar/wk4/art05.htm.
Washington, DC: Author.
Bureau of Labor Statistics. (2003). hnjury, illness andfatality data. Retrieved July 23, 2005 from
http ://www.bls.gov/iif/home.htm. Washington, DC: Author.
Business for Social Responsibility. (2006). About BSR. Retrieved February 9, 2006 from
http ://www.bsr.org/Meta/About/index. cfm. San Francisco, CA: Author.
Canadian Centre for Occupational Health and Safety. (1998). Rotational .\lips~ ork. Retrieved
October 8, 2005 from http ://www.ccohs.ca/oshanswers/work~scheduleshitrhm.
Hamilton, Ontario: Author.
Cellar, D.F., Nelson, Z.C., Yorke, C.M. & Bauer, C. (2001). The five factor model and safety in
the workplace: Investigating the relationship between personality and accident
involvement. Journal of Prevention and Intervention in the Community, 22(1), 43-52.
Cellar, D.F., Yorke, C.M., Nelson, Z.C. & Carroll, K.A. (2004). Relationships between five
factor personality variables, workplace accidents and self efficacy. Psychological Reports,
Cellier, J.M., Eyrolle, H. & Bertrand, A. (1995). Effects of age and level of work experience on
occurrence of accidents. Perception and2~otor Mills//, 80, 931-940.
Centers for Disease Control and Prevention. (1 992). Prinzciples of epidemiology: An introdurction
to applied' epidemiology and' biostatistics (2n ed.). Atlanta, GA: Author.
Centers for Disease Control and Prevention. (1999, June 11). Achievements in public health,
1900-1999: Improvements in workplace safety, -United States, 1900-1999. MWdWR
Weekly, 48(22), 461-469.
Cherry, N., Parker, G., McNamee, R., Wall, S., Chen, Y. & Robinson, J. (2005). Falls and
factures in women at work. Occupational M~edicine, 55(4), 292-297.
Clark, S.P., Rockett, J.L., Sloan, D.M., Aiken, L.H., (2002). Organizational climate, staffing and
safety equipment as predictors of needlestick injuries and near misses in hospital nurses.
American Journal oflnfection Control, 30(4), 207-216.
Cohen, M., Village, J., Ostry, A.S., Ratner, P.A., Cvitkovich, Y. & Yassi, A. (2004). Workload
as a determinant of staff injury in intermediate care. International Journal of Occupational
Environmental Health, 10(4), 375-383.
Cohen-Bendahan, C.C.C., Buitelaar, J.K., van Goozen, S.H.M., Orlebeke, J.F. & Cohen-
Kettenis, P.T. (2005). Is there an effect of prenatal testosterone on aggression and other
behavioral traits? A study comparing same sex and opposite sex twin girls. Hormones and'
Behavior, 47, 230-237.
Colbert, A.E., Mount, M.K., Witt, L.A., Harter, J.K.& Barrick, M.R. (2004). Interactive effects
of personality and perceptions of the work situation on workplace deviance. Journal of
Applied Psychology, 89(4), 599-609.
Cornell University. (2005). The triangle factory fire. Retrieved October 9, 2005 from
http://www.ilr. cornell. edu/trianglefire/. Ithica, NY: Author.
Costa, P.T.J. & McCrae, R.R. (1992). Revised NEO Personality Inventory (NEO-PI-R) and NEO
Five Factor Inventory: Professional manual. Lutz, FL: Psychological Assessment
Cox, A.M. (2003). I am never lonely: A brief history of employee personality testing.
Psychology Issue, 21, Retrieved February 9, 2006 from
http://www. stayfreemagazine. org/archives/21 /personality testing. html. Brooklyn, NY:
Craske, S. (1968). A study of the relation between personality and accident history. British
Journal of2~edical Psychology, 41, 399-404.
Das, T.K. & Teng, B.-S. (2004). A risk based view of trust: A conceptual framework. Journal of
Business and Psychology, 19 (1), 85-116.
Davids, A. & Mahoney, J.T. (1957). Personality dynamics and accident proneness in an
industrial setting. Journal ofApplied Psychology, 41(5), 303-306.
Dembe, A.E. (2001). The social consequences of occupational injuries and illnesses. American
Journal oflndustrial M~edicine, 40 (4), 403-417.
Dement, J.M., Epling, C., Ostbye, T., Pompeii, L.A.. & Hunt, D.L. (2004). Blood and body fluid
exposure risk among health care workers: Results from Duke health and safety surveillance
system. American Journal oflndustrial M~edicine, 46(6), 637-648.
Edlich, R.F., Winters, K.L., Hudson, M.A., Britt, L.D. & Long, W.B. (2004). Preventionof
disabling back injuries in nurses by the use of mechanical patient lift systems. Journal of
Long Term Effects of2\~edicalImplants. 14(6), 521-533.
Engkvist, I., Hjelm, E.W., Hagberg, M,., Menckel, E. & Elenvall, L. (2000). Risk indicators for
reported over-exertion back injuries among female nursing personnel. Epideemiology,
Findoff, M.J., McGovern, P.M., Wall, M., Gerberich, S.G. & Alexander, B. (2004). Risk factors
for work related violence in a health care organization. Injury Prevention, 10(5), 296-302.
Fine, B.J. (1963). Introversion-extraversion and motor vehicle driver behavior. Perception and'
Motor .\ills/, 12, 95-100.
Frank, A. L. (2000). Injuries related to shiftwork. American Journal of Preventive M~edicine, 18
Fransen, M., Woodward, M., Norton, R., Coggan, C, Dawe, M & Sheridan, N. (2002). Risk
factors associated with the transition from acute to chronic occupational back pain. Spine,
Fordyce, W.E., Bigos, S.J., Batti'e, M.C. & Fisher, L.D. (1992). MMPI scale 3 as a predictor of
back injury report: What does it tell us? The Clinical Journal ofPain, 8, 222-226.
Frone, M.R. (1998). Predictors of work injuries among employed adolescents. Journal of
Applied Psychology, 83(4), 565-576.
Garrity, M.J. (2003). Personality, demographics and job relevant risk as predictors of accident
outcomes in the nuclear industry. Dissertation Abstracts International, 65(1-B), 469.
Geller, E.S. (2004, May). Psychology of safety: personality's powerful role in safety.1Industrial
Safety and Hygiene News. Retrieved June 26, 2005 from
http:.//www.i shn. com/CDA/Arti cl elnform ati on/feature s/BNP F features Item/0,2 162, 124
687,00.html. Blacksburg, VA: Author.
Geller, E. S. & Wiegand, D.M. (2005). What 's your safety IQ? Personality traits and person
states related to injury prevention. Retrieved July 17, 2005 from
http://www. safetyperformance. com/AS SE%20PDC%20Paper%202004%20FINAL .doc.
Blacksburg, VA: Author.
Gimeno, D., Felknor, S., Burau, K.D. & Delclos, G.L. (2004). Organizational and occupational
risk factors associated with work related injuries among public hospital employees in
Costa Rica. Occupational and Environmental M~edicine, 62(5), 337-343.
Glanz, K., Rimer, B. K. & Lewis, F.M. (2002). Health behavior and health education: Theory,
research and practice (3rd ed.). Jossey-Bass: San Francisco.
Goldman, R.H., Jarrard, M.R., Kim, R., Loomis, S. & Atkins, E. (2000). Prioritizing back injury
risk in hospital employees: Application and comparison of different injury rates. Journal
of Occupational and Environmental M~edicine, 42(6), 645 -652.
Goodstein, L.D. & Lanyon, R.I. (1999). Applications of personality assessment in the workplace:
A review. Journal of Business and Psychology, 13(3), 291-322.
Green, L.W. (2004). The Procede-Proceed model ofhealth program planning & evaheation.
Retrieved March 21, 2004, from http ://1green.net/precede.htm. San Francisco, CA: Author.
Green, L.W., Richard, L., & Potvin, L. (1996). Ecological foundations of health promotion.
American Journal of Health Promotion, 10(4), 270-28 1.
Haight, F. (2001). Accident proneness: History ofan idea. Institute of Transportation Studies,
University of California. Retrieved July 17, 2005 from
http://www.its.uci. edu/its/publications/papers/WP-0 1-4.pdf. Irvine, CA: Author.
Hair, J.F., Anderson, R.E., Tatham, R.L. & Black, W.C. (1998). M\~ultivariate data analysis (5th
ed.). Prentice Hall: Upper Saddle River, NJ.
Hansen, C.P. (1988). Personality characteristics of the accident involved employee. Journal of
Business and Psychology, 2(4), 346-365.
Harrell, W.A. (1995). Factors influencing involvement in farm accidents. Perception and Motor
Mi.1l/1, 81, 592-594.
Hemingway, M.A. & Smith, C.S. (1999). Organizational climate and occupational stressors as
predictors of withdrawal behaviours and injuries in nurses. Journal of Occupational and
Or ganizational Psychology, 72, 285-299.
Hogan, R., Hogan, J. & Roberts, B.W. (1996). Personality measurement and employment
decisions: Questions and answers. American Psychologist, 51(5), 469-477.
Hogan, R., Johnson, J. & Briggs, S. (Eds.).(1997). Handbook ofpersonality psychology.
Academic Press, Inc.: San Diego.
Hurtz, G.M. & Donovan, J.J. (2000). Personality and j ob performance: the big five revisited.
Journal ofApplied Psychology, 85(6), 869-879.
International Labour Office. (2003). Ethical issues in workplace drug testing in Europe. Geneva:
Iverson, R.D. & Erwin, P.J. (1997). Predicting occupational injury: The role of affectivity.
Journal of Occupational and Organizational Psychology, 70, 1 13-128.
James, L.R. & Mazerolle, M.D. (2002). Personality in work organizations. Sage Publications:
Thousand Oaks, CA.
Janicak, C.A. (1996). Predicting accidents at work with measures of locus of control and job
hazards. Psychological Reports, 78, 115-121.
Jelip, J., Matthew, G.G., Yusin, T., Dony, J.F., Singh, N., Ashaari, M., Lajanin, N., Shanmuga-
Ratman, C., Yusof Ibrahim, M. & Gopinath, D. (2004). Risk factors of tuberculosis among
health care workers in Sabah, Malaysia. Tuberculosis, 84(1-2), 19-23.
Johnston, J.J. (1995). Occupational injury and stress. Journal of Occupational and
Environmental M~edicine, 37(10), 1199-1203.
Jones, J.W. & Wuebker, L. (1985). Development and validation of the safety locus of control
scale. Perceptual and Motor .\d//, 61, 151-161.
Judge, T.A., Heller, D. & Mount, M.K. (2002). Five-factor model of personality and j ob
satisfaction: A meta-analysis. Journal ofAppliedPsychology, 87(3), 530-541.
Judge, T.A., Martocchio, J.J. & Thoresen, C.J. (1997). Five-factor model of personality and
employee absence. Journal ofApplied Psychology, 82(5), 745-755.
Kerr, M. S., Frank, J. W., Shannon, H. S., Norman, R. W. K., Wells, R. P., Neumann, P. &
Bombardier, C. (2001) Biomechanical and psychosocial risk factors for low back pain at
work. American Journal of Public Health, 91(7), 1069-1075.
Kirkcaldy, B.D., Trimpop, R. & Cooper, C.L. (1997). Working hours, job stress, work
satisfaction, and accident rates among medical practitioners and allied personnel.
International Journal of Stress Management, 4 (2), 79-87.
Kirshchenbaum, A., Oigenblick, L. & Goldberg, A.L. (2000). Well being, work environment
and work accidents. Social Science & M~edicine, 50 (5), 631-639.
Lajunen, T. (2001). Personality and accident liability: are extraversion, neuroticism and
psychoticism related to traffic and occupational fatalities? Personality andlndividual
Differences, 31, 1365-1373.
Leigh, J.P., Markowitz, S., Fahs, M. & Landrigan, P. (2000). Costs of occupational injuries and
illnesses. University of Michigan Press: Ann Arbor, MI.
Letvak, S. (2005). Health and safety of older nurses. Nursing Outlook, 53(2), 66-72.
Levy, B.S., Wegman, D.H., Baron, S.L. & Sokas, R.K. (Eds.). (2005). Occupational and
environmental health: Recognizing and preventing disease and injury (5th ed.).
Philadelphia: Lippincott, Williams & Wilkins.
Li, J., Wolf, L. & Evanoff, B. (2004). Use of mechanical life device decreased musculoskeletal
symptoms and injuries among health care workers. Injury Prevention, 10(4), 212-216.
Liao, H, Arvey, R.D., Butler, R.J. & Nutting, S. (2001). Correlates of work injury frequency and
duration among firefighters. Journal of Occupational Health Psychology, 6(3), 229-242.
Lipscomb, J., Trinkoff, A., Brady, B. & Geiger-Brown, J. (2004). Health care system changes
and reported musculoskeletal disorders among registered nurses. American Journal of
Public Health, 94(8), 143 1-143 5.
Marcias, D.J.,Hafner,J.,Brillman, J.C. & Tandberg, D. (1996). Effect of time of day and duration
into shift on hazardous exposure to biological fluids. Academy of Emergency M~edicine,
Markon, K.E., Kruger, R.F., Bouchard, T.J. & Gottesman, I.I. (2002). Normal and abnormal
personality traits: Evidence for genetic and environmental relationships in the Minnesota
study of twins reared apart. Journal ofPersonality, 70(5), 661-693.
Matthews, G. & Deary, I.J. (1998). Personality traits. University Press: Cambridge.
Marras, W.S. (2000). Occupational low back disorder causation and control. Ergonomics, 43(7),
Marusic, A., Musek, J. & Gudjonsson, G. (2001). Injury proneness and personality. Nordic
Journal ofPsychiatry, 55(3), 157-161.
McGill University. (2005). About occupational health. Retrieved July 23, 2005 from
http://www.mcgill .ca/occh/about/#WHAT. Montreal, Quebec: Author.
McLeroy, K.R., Bibeau, D., Steckler, A. & Glanz, K. (1988). An ecological perspective on
health promotion programs. Health Education Quarterly, 15(4), 351-377.
McPhaul, K. M. & Lipscomb, J.A. (2004). Workplace violence in health care: Recognized but
not regulated. Online Journal oflssues in Nursing, 9(3). Retrieved September 3, 2005
from: www.nursingworld. org/oj in/topic25/tpc25_6. htm. Silver Springs, MD: Author.
Merriam-Webster Online. (2005). Retrieved September 5, 2005 from http://www.merriam-
webster.com/. Springfield, MA: Author.
Meucke, S. (2004). Effects of rotating night shifts: literature review. Journal of Advanced
Nursing, 50 (4), 433-439.
Molnos, A. (1998). A psychotherapist 's harvest. Retrieved July 17, 2005 from
http ://fox.klte.hu/~keresofl/psyth/a-to-z-entre/cietproneness.html. Debrecen,
Murphy, G.C. (1994). Job satisfaction of and return to work by occupationally injured
employees. PsychologicalReports, 75, 1441-1442.
Myers, D., Silverstein, B. & Nelson, N.A. (2002). Predictors of shoulder and back injuries in
nursing home workers: A prospective study. American Journal ofhidustral2\~edicine, 41,
National Institute for Occupational Safety and Health. (2005). Health care workers. Retrieved
July 23, 2005 from http://www.cdc. gov/ni osh/topi cs/healthcare/. Atlanta, GA: Author.
National Institute for Occupational Safety and Health. (2001). Conspendium ofNIOSH health
care worker research. Retrieved July 23, 2005 from http ://www.cdc.gov/niosh/docs/2003 -
108/pdfs/2003-108.pdf. Atlanta, GA: Author.
National Safety Council. (2003). Estimating the costs of unintentional injuries, 2003. Retrieved
September 5, 2005 from http ://www.nsc.org/lrs/statinfo/estcost.htm. Itasca, 1L: Author.
National Safety Council. (2005). Glossary ofternas. Retrieved October 7, 2005 from
http://www.nsc. org/lrs/glossary.htm. Itasca, IL: Author.
Nicholson, N. (1996). Towards a new agenda for work and personality: traits, self-identity,
"strong" interactionism and change. Applied Psychology: An hIternational Review, 45(3),
Occupational Safety and Health Administration. (2005). Guidelines for nursing homes:
Ergonomics for the prevention of nusculoskeletal disorders. Retrieved March 26, 2005
http:.//www.premi erinc. com/al l/s afety/re source s/b ack~inj ury/downl oad s/F inal_O SHAGui
delines~nursing~homes.pdf. Washington, DC: Author.
Occupational Safety and Health Administration. (2005). OSH act of 1970. Retrieved July 23,
http://www. osha.gov/pl s/oshaweb/owasrch. search~form?p_doc type=0 SHACT&p toc le
vel=0&ptkeyvalue=&p_status=CURRENT Washington, DC: Author.
Ones, D.S., Viswesvaran, C. & Schmidt, F.L. (1993). Comprehensive meta-analysis of integrity
test validation: Finding and implications for personnel selection and theories of job
performance. Journal ofApplied Psychology, 66, 574-583.
Psychological Assessment Resources, Inc. (2005). NEO personality inventory -revised.
Retrieved October 22, 2005 from
http://www3 .parinc. com/products/product. aspx?Producti d=NEO -PI-R. Lutz, FL: Author.
Rabaud, C., Zanea, A., Mur, J.M., Blech, M.F. Dazy, D., May, T., et al. (2000). Occupational
exposure to blood: Search for a relation between personality and behavior. Infection
Control and Hospital Epidemiology, 21 (9), 564-574.
Roberts, B.W. & Hogan, R. (Eds.).(2001). Personality psychology in the workplace.
Washington, D.C.: American Psychological Association.
Roberts, B.W., Walton, K.E. & Viechtbauer, W. (2006). Patterns of mean-level change in
personality traits across the life course: A meta-analysis of longitudinal studies.
Psychological Bulletin, 132(1), 1-25.
Rosecrance, J.C. & Cook, T.M. (1998). Upper extremity musculoskeletal disorder: Occupational
association and a model for prevention. Central European Journal of Occupational and
Environmental M~edicine, 4(3):214-231. Retrieved February 23, 2005 from www.
http://www.fj okk.hu/cej oem/fil es/Volume4/C onVol4No3 .htm#Rosecrance. Budapest,
Sacks, J.J. & Nelson, D.E. (1994). Smoking and injuries: An overview. Preventive M~edicine, 23,
Salgado, J.F. (2002). The big five personality dimensions and counterproductive behaviors.
International Journal of Selection and Assessment, 10(1/2), 117-125.
Salminen, S., Kivimaki, M., Elovainio, M & Vahtera, J. (2003). Stress factors predicting injuries
of hospital personnel. American Journal oflndustrial M~edicine. 44, 32-36.
Salminen, S. & Klen, T. (1994). Accident locus of control and risk taking among forestry and
construction workers. Perception and Motor .\i/A, 78, 852-854.
Santaguida, P.L. & Fernie, G. R. (1998, March). Efficacy of mechanical hifting devices to
minimize loads to the lower back while performing a heavy transfer hift. Paper presented at
the meeting of the North American Congress on Biomechanics, Waterloo, Ontario,
Canada. Retrieved April 20, 2005 from http://asb-biomech. org/onlineab s/NACOB98/1 08/.
Houston, TX: Author.
Schmit, T.L., Pratt, M., & Howze, E. (1995). Policy as intervention: environmental and policy
approaches to the prevention of cardiovascular disease. American Journal of Public
Health, 85(9), 1207-1211.
Seago, J.A. & Faucett, J. (1997). Job strain among registered nurses and other hospital workers.
Journal ofNursing Administra~tion, 27 (5), 19-25.
Sencan, I., Sahin, I., Yildirim, M. & Yesildal, N. (2004). Unrecognized abrasions
andoccupational exposures to blood-borne pathogens among health care workers in
Turkey. Occupational M~edicine, 54(3), 202-206.
Srivastava, S., John, O.P., Gosling, S.D. & Potter, J. (2003). Development of personality in early
and middle adulthood: Set like plaster or persistent change? Journal ofPersonality and'
Social Psychology, 84(5), 1041-1053.
Stokols, D. (1996). Translating social ecological theory into guidelines for community health
promotion. American Journal of Health Promotion, 10(4), 282-298.
Smith, D.R., Wei, N., Kang, L. & Wang, R.S. (2004). Musculoskeletal disorders among
professional nurses in mainland China. Journal of Professional Nurses, 20(6), 390-395.
Smith, G.S., Wellman, H.M., Sorock, G.S., Warner, M., Courtney, T.K., Pransky, G.S. &
Fingerhut, L.A. (2005). Injuries at work in the US adult population: Contributions to the
total injury burden. American Journal of Public Health, 95 (7), 1213-1219.
Strong, L.L. & Zimmerman, F.J. (2005). Occupational injury and absence form work among
African American, Hispanic and non-Hispanic white workers in the national longitudinal
survey of youth. American Journal of Public Health, 95 (7), 1226-1232.
Swaen, G.M.H., van Amelsvort, L.P.G.M., Bultmann, U., Slangen, J.J.M. & Kant, I.J (2004).
Psychosocial work characteristics as risk factors for being injured in an occupational
accident. Journal of Occupational and Environmental M~edicine, 46 (6), 52 1-527.
Tarasuk, V. & Eakin, J.A. (1994). Back problems are for life: Perceived vulnerability and its
implications for chronic disability. Journal of Occupational Rehabilitation, 4 (1), 55-64.
Taylor, G., Easter, K. & Hegney, R. (2004). Enhancing occupational safety and' health. Elsevier
Butterworth-Heinemann: Oxford, England.
Taub, J.M. (1998). Eysenck's descriptive and biological theory of personality: A review of
construct validity. hIternational Journal ofNeuroscience, 94(3-4), 145-197.
Thoms, P. & Venkataraman, R.R. (2002). Relation of managers' personality to accident and
injury rates. Psychological Reports, 91, 1107-1115.
Triandis, H.C. & Suh, E. M. (2002). Cultural influences on personality. AnnualReview of
Psychology, 53, 133-160.
Trimpop, R., Kirkcaldy, B., Athanasou, J. & Cooper, G., (2000). Individual differences in
working hours, work perceptions and accident rates in veterinary surgeons. Work &
Stress, 14(2), 181-188.
Trinkoff, A.M., Brady, B. & Nielsen, K. (2003). Workplace prevention and musculoskeletal
injuries in nurses. Journal ofNursing Administration, 33(3), 153-158.
U.S. Supreme Court. (1971). Griggs vs. Duke Power, 401 U.S. 424. Retrieved February 9, 2006
from http://caselaw.1p. findlaw. com/scripts/getcase .pl?court=US&vol=40 1&invol=424.
Washington, DC: Author.
Venning, P.J., Walter, S.D. & Stitt, L.W. (1987). Personal and job-related factors as determinants
of incidence of back injuries among nursing personnel. Journal of Occupational M~edicine,
Vollrath, M., Landolt, M.A. & Ribi, K. (2003). Personality of children with accident-related
injuries. European Journal ofPersonality, 17, 299-307.
Watterson, L. (2004). Monitoring sharps injuries: EPINet surveillance results. Nursing Standard~~ddddddddd~~~~
Wilburn, S.Q. & Eijkemans, G. (2004). Preventing needlestick injuries among health care
workers: a WHO-ICN collaboration. hIternational Journal of Occupational and'
Environmental Health, 10(4), 45 1-456.
Wiggins, J.S. (Ed.). (1996). The five factor-nrodel of personality. The Guilford Press: New York.
Wilder-Smith, A., Teleman, M.D., Heng, B.H., Earnest, A., Ling, A.F. & Leo, Y.S. (2004).
Enterging htfectious Diseases, 11(7), 1142-1145.
Wuebker, L.J. (1986). Safety locus of control as a predictor of industrial accidents and injuries.
Journal of Business and Psychology, 1(1), 19-30.
Hilary Stevens Morgan received her bachelor of science in nursing degree from Vanderbilt
University in 1979. She began her nursing career as a staff nurse in the labor and delivery unit at
University Medical Center (now Shands Jacksonville) in Jacksonville, FL.
In 1983, Ms. Morgan earned her master's degree in maternal child nursing from Boston
College. During this time, she was also educated as a Women's Health Nurse Practitioner
(WHNP) for which she gained certification in 1984. She began working as a WHNP at the Clay
County Health Department, Green Cove Springs, FL where she also practiced as a public health
Ms. Morgan j oined the Nurse Midwifery practice at University Medical Center in 1985,
working as a Women's Health Nurse Practitioner. In 1987 she completed the Nurse Midwifery
core courses through the University of Florida College of Nursing and became a Certified Nurse
Midwife (CNM). Ms. Morgan was appointed Administrative Director of the Nurse Midwifery
practice in 1990 and served in that post for 7 years. During this time, she also received an
appointment as clinical faculty for the University of Florida College of Nursing. In 1997 she left
University Medical Center and the University of Florida, and began working in a private
OB/GYN practice in Ocala, FL, where she remains today. She has held an adjunct faculty
position with Jacksonville University School of Nursing since 2002.
Ms. Morgan is a Captain in the Nurse Corps of the United States Navy (Reserve
Component). She currently serves as Officer-in-Charge of Operation Health Support Unit
Jacksonville, Detachments N and R, in Jacksonville, FL. Her military honors include the Navy
Achievement Medal and the Navy Commendation Medal. She is an active member of the Naval
Reserve Association (NRA) and the Association for Military Surgeons of the United States
Ms. Morgan is immediate past chair of the North Central FL chapter of the American
College of Nurse Midwives (ACNM). She is also a member of the Association of Women' s
Health, Obstetric and Neonatal Nurses (AWHONN), the Florida Nurses Association (FNA), and
Sigma Theta Tau International Society for Nurses. She holds certifications as a CNM from
ACNM and as a WHNP from AWHONN.