You are on page 1of 8

160

Work Stress in Aging Police Officers

Gershon et al

Work Stress in Aging Police Officers


Robyn R. M. Gershon, MHS, DrPH
Susan Lin, MPH
Xianbin Li, MHS, PhD

Data are sparse regarding the impact of psychosocial work stress on


the health and well-being of aging workers, even for employees working
in high-stress occupations, such as law enforcement. To improve our
understanding of this issue in older workers, we assessed and characterized work stress, coping strategies, and stress-related health outcomes
in a sample of police officers aged 50 years and older (n 105). The
most important risk factors associated with officers perceived work stress
were maladaptive coping behaviors (eg, excessive drinking or problem
gambling) (odds ratio [OR], 4.95; 95% confidence interval [CI], 2.11
to 11.6) and exposure to critical incidents (eg, shootings) (OR, 3.84;
95% CI, 1.71 to 8.65). In turn, perceived work stress was significantly
associated with anxiety (OR, 6.84; 95% CI, 2.81 to 16.65), depression
(OR, 9.27; 95% CI, 3.81 to 22.54), somatization (OR, 5.74; 95%
CI, 2.47 to 13.33), posttraumatic stress symptoms (OR, 2.89; 95% CI,
1.29 to 6.47), symptoms of burnout (OR, 5.93; 95% CI, 2.54 to
13.86), chronic back pain (OR, 3.55; 95% CI, 1.57 to 8.06),
alcohol abuse (OR, 3.24; 95% CI, 1.45 to 7.22), and inappropriately
aggressive behavior (OR, 4.00; 95% CI, 1.34 to 11.88). These data
suggest that older workers in high-stress jobs may be at increased risk for
work stressrelated health problems, especially if they rely on risky health
behaviors to cope with stress. Given the size of the rapidly aging US
workforce and the likelihood that many are employed in high-stress jobs,
interventions are urgently needed to address this emerging public health
issue. (J Occup Environ Med. 2002;44:160 167)

From the Mailman School of Public Health, Columbia University.


The points of view expressed by the authors do not necessarily represent the official positions or
policies for the National Institute of Justice or the US Department of Justice.
Address correspondence to: Dr Robyn R.M. Gershon, Mailman School of Public Health, Columbia
University, 600 W. 168th Street, 4th Floor, New York, NY 10032; rg405@columbia.edu.
Copyright by American College of Occupational and Environmental Medicine

n line with overall demographic


trends in the United States, unprecedented changes in the US workforce
will occur over the next two decades
as the number of workers in the 45to 64-year age range increases at a
rate three times faster than any other
age workgroup.1 By 2012, the first
wave of post-World War II baby
boomers will reach retirement age;
by 2027, the largest number of individuals in this nations history will
have reached 65 years of age.2 As
these workers increasingly exit the
workforce, severe labor shortages,
especially in certain growth sectors
(such as health care, law enforcement, and other service industries),
will result.3
Some employers are trying to
counter this trend by encouraging
older employees to postpone retirement, and this concept has in turn led
to an increased interest in how best
to accommodate aging workers so
that they may safely and effectively
remain on the job.4 This is especially
important for aging workers with
jobs that place them at high risk for
occupational injury or exposure, including exposure to workplace stressors. Aging workers employed in
high-stress occupations may be at
risk for increased morbidity and
mortality rates related to current
stress, coupled with the cumulative
effects of stress-related behaviors,
such as smoking, alcohol abuse, lack
of exercise, and poor nutrition.5 Although little is known about the work
stress process across the life span,
there is some evidence that less effective coping mechanisms are increasingly applied with advancing
age, and this could make stressful
jobs even riskier for older workers.6
Also, workers already at risk because

JOEM

Volume 44, Number 2, February 2002

of the long-term effects of stress may


be especially vulnerable to retirement-related stress, especially if they
have not adequately planned for this
stage of their lives.7
Certain aspects of work (such as
high demand/low control, shiftwork,
and frequent contact with the general
public) are known to be highly
stressful, and jobs with these characteristics are often referred to as
high-stress jobs.8,9 Policing, in
particular, has long been considered
one of the most stressful occupations. Over 25 years ago, Kroes and
Hurrell10 and others noted that exposure to line-of-duty critical incidents
and certain aspects of police organizations were especially stressful for
officers.1113 Examples of stressful
management characteristics associated with policing include complex
and bureaucratic department organizational structures, poor communication among divisions, lack of diversity, rigidity regarding policies and
procedures, lack of opportunity for
advancement, poor supervision, and
various job/task factors (such as
workload).14,15 In terms of critical
incidents, one of the most stressful
experiences in policing, not surprisingly, is the death or injury of another officer in the line of duty. 12,16,17 Other highly stressful
critical incidents include involvement in a hostage crisis, the arrest of
a violent perpetrator, and being investigated by the internal affairs
division.18 20
In keeping with current work
stress/health models, these and other
types of stressors may potentially
lead to adverse health outcomes (ie,
stressors 3 perceived stress 3
stress outcomes).2123 However, research has shown that this relationship may be modified by certain
types of coping styles, in particular,
by the application of problemsolving coping strategies (eg, making
a plan of action and following
through).24,25 Alternatively, coping
responses may be maladaptive, and,
in law enforcement, such behaviors
(eg, alcohol abuse, cigarette smok-

ing) are common. 26,27 Various


stress-related health outcomes (eg,
abnormal norepinephrine levels, alcoholism, heart disease) are also reportedly common in policing.28,29
For example, in one retrospective
study of police mortality, officers
had significantly elevated ageadjusted standardized mortality ratios related to ischemic heart disease
(a compensable disease in police
work), and the risk of mortality from
arteriosclerotic heart disease tended
to increase with increasing years of
employment.30 Interestingly, a recent study by Franke and colleagues
noted that the rate of myocardial
infarcts in police officers increased
in the initial postretirement period.31
In addition to physical health outcomes, researchers also report a high
incidence of adverse-related psychological outcomes in this population
of workers; for instance, age-adjusted suicide rates are significantly
higher in police officers compared
with other workers, and one study
noted that police officers had the
third highest suicide rate among 130
occupations.32,33
The effects of work stress in older
law enforcement personnel may potentially represent a public health
problem of considerable magnitude,
given the sizable and rapidly aging
workforce; as of October 1999, there
were nearly 900,000 law enforcement personnel, including 637,551
sworn (ie, with power to arrest and
licensed to carry a gun) and 261,567
civilian employees.34 In addition to
the public health costs associated
with work stress outcomes in this
population, the impact on police departments may also be considerable.
For example, work stress may result
in high rates of job dissatisfaction,
low morale, poor productivity, high
absenteeism, high turnover, early retirement, high accident rates, poor
public relations, and a high incidence
of lawsuits.35,36 Troubled police departments may have serious difficulties in recruiting new officers. Importantly, work stress may also be
associated with inappropriate and de-

161

viant behaviors on and off the job,


thereby jeopardizing officers; their
families, friends, and coworkers; and
the general public.36 As part of a
large-scale study, referred to as
Project SHIELDS, which examined
the relationship between perceived
work stress and adverse stressrelated outcomes in a sample of police officers, we separately analyzed
a subset of data from officers aged
50 years and older to characterize the
relationship between stress and
health in older officers.37

Methods
Sample
A convenience sample of sworn
police officers (ie, licensed to carry a
weapon and with arrest powers) employed by a large urban police department was surveyed. Police officers were recruited to voluntarily
complete a self-administered anonymous questionnaire during roll call,
which took place at each shift for the
13 departmental districts throughout
the city. From a total of 1106 responding officers, we selected data
from a subset of 105 officers aged 50
years and older. All procedures involving human subjects had prior
approval by the Johns Hopkins University School of Public Health Institutional Review Board.

Study Questionnaire
We designed a five-page, selfadministered survey to collect demographic and psychosocial data on
four major study constructs: (1)
stressors, (2) perceived stress, (3)
coping mechanisms, and (4) stressrelated health outcomes. To simplify
administration, the questionnaire was
aimed at a tenth-grade reading level
and took approximately 20 minutes
to complete. Copies of the study
questionnaire, coding information,
and scale psychometrics may be obtained from the senior author
(RRMG).

162

Measurement
Responses to most items used a
four-point Likert-type scale (eg,
from strongly agree to strongly
disagree).38 All new or revised
scales underwent psychometric evaluation, including Cronbachs alpha.39 The questionnaire included
items on the following:
demographic information: age,
gender, race/ethnicity, marital status, education, tenure, and history
of military service
stressors (ie, sources of stress at
work): 14 items, derived from
Beehr et als police stress questionnaire,40 grouped into four categories: (1) critical incidents (eg,
shooting someone, attending a police funeral, responding to a chemical spill); (2) poor cooperation
(eg, there is a good and effective
cooperation between units) [reverse scored]; (3) perceived inequality, (eg, promotions in the
department are tied to ability and
merit) [reverse scored]; and (4)
organizational structure (eg, the
administration supports officers
who are in trouble [reverse
scored]
perceived work stress: 11-items on
perceived (or felt) job stress (eg,
I feel negative, futile or depressed
about work, My interest in doing fun activities is lowered because of my work, I have difficulty concentrating on my job, I
feel tired at work even with adequate sleep,) from a well-characterized scale previously adapted
from the National Institutes for
Occupational Safety and Health
work stress scale41,42
coping strategies: 10 items total: 3
on problem-solving, 2 on emotionfocused, 1 on avoidance strategies,
and 4 items on negative behaviors
(eg, drinking alcohol), adapted
from scales developed by Beehr et
al and Billings and Moos40,43
stress-outcomes, categorized into
three health outcome subsets: (1)
psychological, including anxiety
(4 items), depression (9 items),
somatization (6 items), burnout (4
items), and posttraumatic stress

Work Stress in Aging Police Officers

Gershon et al

TABLE 1
Demographics of Police Officers*
Characteristic
Mean age (yrs)
Gender
Male
Female
Race
White
Nonwhite
Education
High school
Some college
College/graduate school
Tenure on the police force (yrs)
Current rank
Officer
Supervisor
Marital status
Married
Unmarried
Military service
Yes
No

% or Range

53

50 67

103
2

98.1
1.9

92
13

87.6
12.4

16
12
37
28

15.2
49.5
35.3
2 44

46
59

43.8
56.2

82
23

78.1
21.9

67
38

63.8
36.2

* n 105.

disorder (PTSD) symptoms (3


items)44,45; (2) physical, including
an eight-item health symptom
scale (eg, migraines, high blood
pressure, chronic back problems);
and (3) behavioral, including
smoking (1 item), alcohol abuse (3
items), accidents (1 item), and aggressive behavior (4 items).46

Analyses
After data cleaning and editing
procedures, we conducted an array of
descriptive statistics (frequencies,
means, and standard deviations) for
all variables. Then, we factoranalyzed all new scales and determined their reliability using Pearsons coefficient correlation.47 This
was followed by chi-squared tests of
association between stressors and
perceived stress and between perceived stress and health outcomes.
Finally, we developed simple logistic
regression models to determine the
odds ratio (OR) and 95% confidence
intervals (95% CI) for the association between stressors and perceived
work stress and between perceived
work stress and adverse psychological physical and behavioral out-

comes. To control for possible confounding variables (such as age,


race, tenure in the field, history of
military service), multiple logistic regression equations were constructed.
All variables significant at the univariate level were included in the
final regression models.

Results
Descriptive Statistics
Response rate and demographic
characteristics. Of 1880 police officers potentially eligible to participate
in the larger survey, 1106 completed
the anonymous questionnaire, resulting in a 70% response rate. From the
1106 returned questionnaires, a total
of 105 usable questionnaires were
from officers aged 50 years or older.
This rate represents 9.4% of the total
sample, which correlates well with
the police departments statistics indicating that 10% to 12% of the
sworn police force falls in this age
range. The sample of older officers
had a mean age of 53.5 years and
was predominantly composed of
white men. In general, the demographic profile of the older subset of

JOEM

Volume 44, Number 2, February 2002

163

TABLE 2
Critical Incidents in Policing
% Reporting High
Emotional Affect
From Incident

Type of Incident
Attending a police funeral
Experiencing a needlestick or other exposure to
bloody and body fluids
Being the subject of an internal affairs investigation
Shooting someone
Personally knowing a victim
Making a violent arrest
Being involved in a hostage situation
Responding to a bloody crime scene
Responding to a call related to a chemical spill

officers was similar to that of officers 49 years of age and younger,


except that the older group included
fewer women (15% of the departments police officers are female; in
our sample, only 1.9% were female).
Also, our sample of older officers
had fewer racial/ethnic minority officers than the department as a whole
(eg, 12.4% in our sample vs 35% for
the entire department). Table 1 illustrates the demographic characteristics of the sample of 105 officers.
Stressors (sources of stress at
work). All of our multiple-item measures, including the stressor scales,
had acceptable levels of internal reliability, with Cronbach alphas of
0.75 or greater. The mean score
(with a range of 1.00 to 4.00, with
4.00 being the highest score) for each
the organizational stressors scales
were as follows: lack of cooperation
between units 3.50, inequities at
work 3.23, and poor organizational structure 2.81. The mean
score for the critical incidents scale
was 2.40 (of a possible score of 3.00
as the highest). Table 2 presents the
reported frequency of distress related
to examples of various critical incidents. As expected, the critical incident that most officers (77%) found
especially upsetting was attending a
fellow officers funeral, followed by
experiencing needlestick injuries
(60%).
Perceived stress. The perceived
work stress scale (alpha 0.85) had

77.5
60.0
50.0
47.5
41.8
23.8
17.7
13.9
12.7

an overall mean score of 1.74 (SD,


0.4), and a range of 1.00 to 3.27 of a
possible range of 1.00 to 4.00 (4.00
being the highest). For the analyses,
we assigned officers to either the
high or low stress category (depending on whether their stress score fell
above or below the mean score).
Frequency scores were particularly
high for certain items on the scale;
for example, 64% reported that they
sometimes or more often felt negative, futile, or depressed about
work; 78% frequently felt that they
were not as efficient at work as I
should be; 66% often felt physically, emotionally, and spiritually depleted; 46% sometimes felt uncaring about the problems and needs of
the public when I am at work; and
69% reported that they sometimes
agreed with the statement, When I
ask myself why I get up and go to
work, the only answer that occurs to
me is that I have to. A majority of
officers (55%) reported that they
would likely look for another job
within the coming year.
Coping. Officers frequently relied
on the following problem-solving
coping strategies to deal with the
stress of their work: talking to family
or professionals when they felt
stressed (85%), drawing on past experiences (50%), and making a plan
of action and following through
(47%). Officers also reported using
emotion-focused strategies, ie, praying (81%) and relying on their faith

in God (32%). Examples of commonly used negative or avoidance


strategies included hanging out at
bars with fellow officers (30%); yelling at others, such as family members (55%); smoking more than
usual (30%); gambling (17%); and
acting as if nothing was bothering
them when they were feeling
stressed (26%) (passive withdrawal).

Health Outcomes
Psychological symptoms. Of a
possible range of 1.00 to 3.00 (never,
sometimes, often), the mean scores
for the three stress-related psychological scales were as follows: anxiety 1.28 (range, 1.00 to 2.25),
somatization 1.43 (range, 1.00 to
2.33), and depression 1.51 (range,
1.00 to 2.56). The symptoms most
often reported by officers were low
energy (87%), feeling blue (79%),
headaches and pressure in the head
(58%), no interest in things (55%),
self-blame (53%), pains or pounding
in the chest (53%), loss of sexual
interest (48%), and stomach pains
(43%). Seven percent of the officers
reported that they sometimes thought
about ending their life. Symptoms of
PTSD were common; 27% of the
respondents reported that they had
intrusive or recurrent thoughts, memories, or dreams about distressing
work events; 26% avoided anything
related to the stressful event; and
26% felt detached from people and
activities that they believe were related to the stressful event. Symptoms of burnout were also frequent;
for instance, 40% stated that they felt
burned out from the job, 31% were
on automatic pilot most of the
time, 12% treat the public as if
they were impersonal objects, and
13% stated that they are at the end
of their rope.
Physical symptoms. The most
commonly reported physical symptoms included chronic low back pain
(45%), high blood pressure (42%),
foot problems (32%), heart disease
(16%), migraines (14%), and chronic
insomnia (13%).

164

Work Stress in Aging Police Officers

Univariate

Multivariate

Variable

OR

95% CI

OR

95% CI

Critical incidents
Poor cooperation
Maladaptive coping

3.84
2.23
4.95

1.71 8.65
1.01 4.96
2.1111.6

3.71
NS
5.35

1.2610.9
1.7516.35

* OR, odds ratio; CI, confidence interval; NS, not significant.

Not significant at the univariate level, perceived inequity, organizational structure.

Adjusted for demographic variables.

Relationship Between Stressors


and Perceived Stress
Of the four categories of stressors
measured, two were significantly associated with perceived work stress:
lack of cooperation within the department and exposure to critical incidents (Table 3). Other aspects of
policing (such as perceived inequities and rigid organizational structure) were not significantly associated with perceived work stress. The
relationship between coping and perceived stress was significantly corre-

Gershon et al

ing variables, all main effects remained significant (Table 4).

TABLE 3
Adjusted OR of Stress by Sources of Stress at Work*

Health risk behaviors. The officers


also reported several behaviors previously linked to work stress.48,49,50
For instance, many officers reported
that they currently smoke tobacco
products (68%), 13% worried or felt
guilty about their alcohol consumption, 38% reported that they sometimes drank more than they had
planned, and 13% stated that they
sometimes did not remember what
happened when they were drinking.
Because work stress may also be a
risk factor for workplace accidents,
we asked the officers about these
accidents, and nearly 6% stated that
they had experienced a serious injury
on or off the job within the previous
6 months. Similarly, we asked about
aggressive behavior, and 17% of the
respondents admitted that they sometimes smashed things to relieve
their stress. Officers also acknowledged sometimes getting physical,
ie, by pushing, shoving, grabbing,
and hitting their fellow officers
(6%); spouse or significant other
(5%); children (6%); or pets (10%).

lated; officers who reported using


maladaptive coping strategies to deal
with stress were nearly five times
more likely to report perceived work
stress than officers who relied on
problem-solving coping strategies
(ie, making a plan of action and
following through, or seeking professional help/guidance) (OR, 4.95;
95% CI, 2.11 to 11.6). Each of the
stressors was entered into multivariate models with potential confounders, such as demographic variables
(ie, race, education, tenure), and critical incident exposure and maladaptive coping behaviors remained significantly associated with work
stress (Table 3).

Relationship Between Perceived


Stress and Health Outcomes
As Table 4 indicates, perceived
work stress was significantly associated with several adverse health and
behavioral outcomes. For instance,
officers reporting higher levels of
work stress were significantly more
likely than those with lower levels of
stress to report symptoms of anxiety,
depression, somatization, PTSD, and
burnout. Higher levels of work stress
also correlated with negative physical health symptoms (such as chronic
back pain and foot problems) and
with risky behaviors, including
symptoms of alcohol dependency
and aggressive behavior. In this sample of officers, perceived work stress
was not associated with high blood
pressure, migraine, or heart disease.
In a multiple regression model that
controlled for demographic and cop-

Discussion
A key finding of this study is that
older officers with higher levels of
work stress are at significant risk of
serious physical, mental, and health
risk problems. Three of four officers
reporting stress also reported symptoms of depression, and nearly one of
every two stressed officers also had
symptoms of PTSD. Of special concern was the finding that all seven of
the officers (7%) reporting suicidal
thoughts also reported higher stress
levels. Officers with higher stress
were also more likely to report risky
health behaviors; 60% of those reporting high stress also reported
problem drinking patterns and nearly
one third reported inappropriately
aggressive behavior.
Our findings generally support
earlier stress research linking work
stress to health problems and health
risk behaviors.51 Our findings also
support and extend other police
stress research. For instance, in addition to several critical incidents commonly reported as stressful, we
found that needlestick injuries were
also highly stressful. Given that 50%
of the officers reported a history of
needlestick injury, this type of exposure may represent a law enforcement problem of some concern, especially because the negative impact
of needlesticks was second only to
attending a police funeral.
In contrast to other police stress
research, we did not obtain the same
level of association between police
stress and various organizational
stressors. For example, although
both organizational structure and inequitable treatment have previously
been identified as important sources
of police work stress, this was not the
case for this older cohort. It may be
that our respondents learned, over
time, to adapt to the management
structure so that it no longer bothered
them, or perhaps officers who were
especially distressed by these aspects
were no longer on the police force.

JOEM

Volume 44, Number 2, February 2002

165

TABLE 4
Adjusted OR of Stress-Related Health Outcomes by Perceived Sources of Stress at Work*
% Reporting

Outcome Variable
Psychological
Anxiety
Depression
Somatization
PTSD
Burnout
Physical
Chronic back pain
High blood pressure
Migraine
Foot problems
Heart disease
Behavior
Alcoholism
Smoking
Accidents
Aggression

Multivariate

Univariate

High
Stress

Low
Stress

OR

95% CI

OR

95% CI

57.8
75.6
71.1
53.0
73.0

16.7
25.0
30.0
28.0
32.0

6.84
9.27
5.74
2.89
5.93

12.81 6.65
3.1822.54
2.4713.33
1.29 6.47
2.5413.86

4.06
10.14
5.60
2.89
6.56

1.4911.04
3.8826.54
2.3513.35
1.28 6.51
2.5117.09

62.8
36.4
18.6
48.8
19.1

32.2
45.0
10.5
19.0
13.6

3.55
NS
NS
4.08
NS

1.57 8.06

4.04
NS
NS
4.09
NS

1.5010.84

60.0
40.0
9.0
27.0

32.0
22.0
3.0
8.0

3.24
NS
NS
4.00

1.457.22

2.76
NS
NS
5.13

1.017.55

1.699.81

1.3411.88

1.6310.23

1.5217.29

* OR, odds ratio; CI, confidence interval; PTSD, posttraumatic stress disorder; NS, not significant.
Adjusted for demographic and coping variables.

The lack of a relationship between


inequitable treatment and perceived
stress may also be explained by the
homogeneity of the sample, as most
of the sample consisted of white
men. We also did not find, as other
studies have shown, an association
between stress and cardiac health
problems. This may be related to the
age of our sample, ie, officers with
serious cardiac and other health
problems might simply opt to retire
and, thus, would not be represented
in the sample. Similarly, we also did
not find a relationship between stress
and accidents, either on or off the
job, probably because accidents were
relatively rare, with only six reported
over a 6-month period. Perhaps with
a larger sample size we would have
had enough power to discern this.
In line with other studies on coping, we also found that negative coping skills were highly significantly
associated with perceived work
stress.52,53 However, in contrast to
other stress studies, poor coping did
not intensify or mediate the relationship between stressor and felt
stress; rather, in this study, poor coping was itself an important stressor.

In other words, instead of being an


antidote to stress, negative coping
behaviors (eg, drinking) were more
likely to be related to the feeling of
being stressed. As Beehr and others
have pointed out, the copingstress
process is conceptually difficult because negative coping behaviors
may, at times and in some circumstances, lessen the feeling of stress at
the same time that they themselves
are sources of stress or strain.54
Beehr suggests that the balance may
tilt in one direction or the other based
on the extent of the behavior (ie, the
degree of drinking). Furthermore,
these behaviors may also be considered as stress outcomes. Thus, the
stress-health model has sometimes
been depicted as a circular process,
ie, stressors 3 stress 3 outcomes 3
stressors. Because research indicates
that some of these negative coping
skills may be adopted early in the
officers careers as part of the socialization process for new police recruits, it is important to consider the
risk factors leading to the initial
adoption of these behaviors.53 Of
course, there is also the possibility
that people with these risk factors

may self-select into the profession. It


is interesting to consider why some
officers adopt and maintain these
negative behaviors, whereas others
switch to more effective coping behaviors over time. Even though negative coping behaviors were not
helpful in reducing perceived stress
in our cohort, and in fact worsened
the feeling of stress, roughly one
third of the officers still primarily
relied on these methods.
Based on our results, we can assume that older workers, as with
younger workers, may similarly benefit from risk reduction strategies
that modify stressors, support effective coping, or both. The primary
prevention strategy for managing
work stress is to remove, reduce, or
somehow modify work stressors.
However, when this may be impracticable or not feasible (eg, in policing, certain critical incident stressors
may be inherent to the job itself),
then effective coping skills become
even more important. Positive coping may be organizationally encouraged through a variety of approaches
(eg, through skills-building workshops). Organizations can also sup-

166

port stress debriefing programs, such


as critical incident debriefing, which
may be helpful in reducing the incidence of PTSD in exposed workers.
This may be an issue of particular
interest in older police officers, because many (25%) of our respondents reported symptoms of PTSD;
conceivably, this condition may be
related to past traumatic events that
occurred before the availability of
critical incident debriefing programs.
It would be of interest in future
studies to determine the effectiveness of various therapeutic approaches for treating work-related
PTSD resulting from critical incident
exposure that occurred in the distant
past.
Several potential limitations of
this study must be addressed. For
example, the cross-sectional design,
although convenient and cost-effective, precludes the determination of
causality. Nevertheless, this is an
effective approach for exploring the
nature of the relationship between
determinants and an important first
step toward more definitive prospective studies. Another potential limitation is the issue of the generalizability of the findings to other aging,
at-risk work groups, or even to other
police departments. Although there
is no reason to suspect that this
population of older officers differs
greatly from officers in other police
departments, it would be helpful to
repeat this study in other cohorts and
in other completely different work
settings, because other older cohorts
may have different health outcomes
and different methods of coping and
may, therefore, require different intervention approaches. Stress studies
on other aging, high-risk workgroups
(such as nursing personnel) are also
urgently needed.
Another potential limitation to
consider is selection bias. For example, if the differences between study
participants and nonparticipants
were related to either stress risk factors or stress outcomes, then both
validity and strength of association
might be affected. For example, if

Work Stress in Aging Police Officers

officers with greater levels of stress


were out on medical leave, then the
sample would underestimate the effect of stress on health. If, on the
other hand, officers with higher levels of stress were more likely to be
motivated to respond, then the effect
might be overestimated. Also, because the cumulative effects of adverse coping, along with stressrelated health problems, can increase
the morbidity and mortality of
stressed workers, the most severely
affected older officers might be underrepresented. Officers who were
absent or on leave or who quit, retired, or died because of their experiences with stress would not be represented at all. In essence, therefore,
our sample represents the survivors, and this also has the potential
to lead to underestimation of the
strength of association between
stress and stress outcomes. Survival
bias might also help to explain the
lack of association in our data between perceived stress and certain
health outcomes (such as heart disease). To address these problems, we
used relative rather than absolute
measures in our analyses, thereby
enhancing the generalizability of
effect.

Conclusions
These results highlight the importance of the development and implementation of interventions that either
moderate stressors or improve coping skills, or both, for aging, highrisk workers. Specific stress reduction interventions that address the
needs and concerns of the older work
population are needed. For instance,
aging workers may benefit not only
from job/task modifications that support their continued and safe employment but also from special programs that enhance their coping
capabilities by encouraging the substitution of negative health habits for
more effective and health-promoting
strategies. Support groups for older,
at-risk workers might be structured
to promote healthy preretirement,
and membership in these groups

Gershon et al

could be maintained well into retirement to serve as a continued source


of social support. An example of this
is the role that police fraternal organizations often play by involving retired officers in various police department events. These programs
help to keep police retirees connected to the department and to one
another.
Obviously, much more research is
needed, not only on the epidemiology of work stress across the life
span but also on effective interventions designed for the older worker.
The urgency and potential seriousness of the situation cannot be overstated, given the imminent and extraordinary changes facing the US
workforce in the coming decades.

Acknowledgments
The original study Project SHIELDS
was funded by the National Institute of Justice. The authors appreciate the editorial assistance of Ms Jane Shipley and the review
comments of Dr Ray Sinclair. A special thank
you to Mr Gary McLhinney, President of the
Fraternal Order of Police, Baltimore Chapter;
Ms Dottie Woods; and Colonel Margaret
Patten for their helpful and generous support
for this project.

References
1. Bureau of Labor Statistics. Occupational
Outlook Handbook. Washington, DC;
1998.
2. Bureau of Labor Statistics. Employment
Projections. Washington, DC; 1998.
3. Judy RW, DAmico C, Geipel GL. Workforce 2020. Hudson Institute Publication;
1997.
4. Beehr TA, Glazer S. Work and non-work
predictors of employees retirement age.
J Vocational Behav. 2000;57:206 225.
5. Pearlin L, Skaff MM. Stress and the life
course: a paradigmatic alliance. Gerontologist. 1996;36:239 247.
6. Aldwin CM. The elders life stress inventory: egocentric and nonegocentric stress.
In: Stephens MAP, Crowther JH, Tennenbaum DL, eds. Stress and Coping in
Later-Life Families. New York: Hemisphere; 1990:49 69.
7. Talaga J, Beehr T. Retirement: a psychological perspective. In: Cooper CL, Roberston I, eds. International Review of
Industrial and Organizational Psychology. New York: Wiley & Sons; 1989:
185208.
8. Sauter SL, Murphy LR. Organizational
Risk Factors for Job Stress. Washington,

JOEM

9.

10.
11.
12.

13.
14.
15.

16.
17.

18.

19.
20.

21.
22.

23.

24.

Volume 44, Number 2, February 2002

DC: American Psychological Association; 1995:1 6.


Muntaner C, Eaton WW, Garrison R.
Dimensions of psychosocial work environmental in a sample of US metropolitan population. Work Stress. 1993;7:351
363.
Kroes WH, Hurrell JJ. Job Stress and the
Police Officer. Washington, DC: US
Government Printing Office; 1975.
Kroes WH. Societys Victim, the Policeman: An Analysis of Job Stress in Policing. New York: Charles C Thomas; 1976.
Finn P, Tomz JE. Developing a Law
Enforcement Stress Program for Officers
and Their Families. Washington, DC:
National Institute of Justice; 1997:119.
Brown JM, Campbell EA. Stress and
Policing, Sources and Strategies. New
York: Wiley & Sons; 1994.
Brown JM, Campbell EA. Sources of
occupational stress in the police. Work
Stress. 1990;4:305371.
Brown JM, Fielding J. Qualitative differences in men and women police officers
experience of occupational stress. Work
Stress. 1993;7:327340.
Violanti JM, Aron F. Ranking police
stressors. Psychol Rep. 1994;75:824
826.
Jermier JM, Gaines J, McIntosh NJ. Reactions to physically dangerous work: a
conceptual and empirical analysis. J Organizational Behav. 1989;10:1523.
Eisenberg T. Job stress and the police
officer: identifying stress reduction techniques. In: Kroes WH, Hurrell JJ, eds.
Job Stress and the Police Officer: Identifying Stress Reduction Techniques.
Washington, DC: US Government Printing Office; 1975:26 34.
Kirschman E. I Love a Cop. New York:
The Guilford Press; 1997.
Horn J. Critical incidents for law enforcement officers. In: Reese J, Horn J, Dunning C, eds. Critical Incidents in Policing. Washington, DC: US Government
Printing Office; 1991:143148.
Everly GS, Smith K. Occupational stress
and its management. Human Stress: Current selected research. 1987;2:235246.
Jaffe DT. The healthy company: research
paradigms for personal and organizational health. In: Sauter SL, Lawerence
MR, eds. Organizational Risk Factors for
Job Stress. Washington, DC: American
Psychological Association; 1995:13 40.
Quick J, Quick JD, Nelson D, Hurrell JJ.
Preventive stress management in organizations. In: Meidenbauer E, ed. Individual Consequences of Stress. Washington,
DC: American Psychological Association; 1997:65 87.
Latack JC, Havlovic SJ. Coping with job
stress: a conceptual evaluation framework for coping measures. J Organizational Behav. 1992;13:479 508.

25. Dewe PJ, Guest DE. Methods of coping


with stress at work: a conceptual analysis
and empirical study of measurement issues. J Organizational Behav. 1990;11:
135150.
26. Violanti JM, Marshall JR, Howe BH.
Stress, coping, and alcohol use: the police
connection. J Police Sci Admin. 1985;13:
106 110.
27. Reese J, Scrivner E. Law Enforcement
Families: Issues and Answers. Washington, DC: US Government Printing Office; 1994;57 66.
28. Vena JE, Violanti JM, Marshall J, Fiedler
R. Mortality of a municipal worker cohort: police officers. Am J Ind Med.
1986;10:383397.
29. Ely D. Organization of cardiovascular
and neurohumoral responses to stress:
implication for health and disease. Ann
NY Acad Sci. 1995;771:594 608.
30. Sardinas A, Miller JW, Hansen H. Ischemic heart disease mortality of firemen
and policemen. Am J Public Health.
1986;76:1140 1141.
31. Franke WD, Cox DF, Schultz DP. Coronary heart disease risk factors in employees Iowas department of public safety
compared to a cohort of the general
population. Am J Ind Med. 1997;31:733
737.
32. Fell RD, Richard WC, Wallace WL. Psychological job stress and the police officer. J Police Sci Admin. 1980;8:139 144.
33. Allens S. Suicide and indirect selfdestructive behavior among police. In:
Reese J, Goldstein H, eds. Psychological
Services for Law Enforcement. Washington, DC: US Government Printing Office; 1986.
34. www.bls.gov/oco/ocos160.htm.
35. Hurrell JJ, Pate A, Kliesmet R. Stress
Among Police Officers. Cincinnati, OH:
National Institute of Occupational Safety
and Health; 1984.
36. Tang TL, Hammontree ML. The effects
of hardiness, police stress, and the life
stress on police officers illness and absenteeism. Public Personnel Management. 1992;21:493510.
37. Gershon RRM, Patten M, Johnson O.
Project shields. In: Proceedings of the
National Symposium on Law Enforcement Families, San Antonio, TX, 1999.
38. Likert R. A technique for measurement of
attitudes. Arch Psychol. 1932;140:155.
39. Cronbach LJ. Coefficient alpha and the
internal structure of tests. Psychometrika.
1951;16:297334.
40. Beehr TA, Johnson LB, Nieva R. Occupational stress: coping of police and their
spouses. J Organizational Behav. 1995;
16:325.
41. Gershon RRM, Karkashian C, Vlahov D,
et al. Compliance with universal precautions in correctional health care facilities.
J Occup Environ Med. 1999;41:181189.

167
42. Hurrell J, Murphy LR. An overview of
occupational stress and health. In: Rom
WM, ed. Environmental and Occupational Medicine. 2nd ed. Boston: Little
Brown; 1982:675 684.
43. Billings AG, Moos RH. The role of
coping responses and social responses in
attenuating the stress of life events. J
Behav Med. 1981;4:139 157.
44. Derogatis LR. Description and Bibliography for SCL-90. Baltimore: Johns Hopkins University Press; 1981.
45. Maslach C, Jackson SE. MBI: Maslach
Burnout Inventory; Manual Research
Edition. 2nd ed. Palo Alto, CA: Consulting Psychologist Press; 1986.
46. Goldenhar LM, Gershon RRM, Mueller
C, Karkasian C, Swanson NA. Psychosocial work stress in female funeral service practitioners. Gender Work Health.
2001;20:1738.
47. Fleiss JL. Statistical Methods for Rates
and Proportion. 2nd ed. New York:
Wiley & Sons; 1981.
48. Jones JW, Boye MW. Job stress and the
employee counterproductively. In: Quick
JC, Murphy LR, Hurrell JJ, eds. Stress &
Well-Being at Work. Washington, DC:
American Psychological Association;
1992:239 251.
49. Parrott AC. Stress modulation over the
day in cigarette smokers. Addiction.
1995;90:233244.
50. Guastello SJ, Gershon RRM, Murphy
LR. Catastrophe model for the exposure
to blood-borne pathogens and other accidents in health care settings. Accid Anal
Prev. 1999;31:739 749.
51. Hurrell J, Murphy L. Psychological job
stress. In: Rom W, ed. Environmental
and Occupational Medicine. Boston: Little, Brown; 1992:675 684.
52. Johnson J, Hall E. Job strain, workplace
social support and cardiovascular disease: a cross sectional study of a random
sample of the Swedish working population. Am J Public Health. 1988;78:1336
1342.
53. Johnson LB, Subramanian G. Culturalization of emotions: police burnout and
work and family violence. Paper presented at Work and Family: Expanding
the Horizons Conference of Business and
Professional Womens Foundation, Center for Working Families at the University of California, Berkeley and the Alfred P. Sloan Foundation, San Francisco,
CA, 2000.
54. Beehr T, McGrath J. The methodology
of research on coping: conceptual, strategic, and operational level issues. In:
Zeidner M, Endler NS, eds. Handbook
of Coping. New York: Wiley & Sons;
1996:65 85.