The Hidden Dangers of Attending Work While Unwell: A Survey Study of Presenteeism Among Pharmacists

Abstract

Presenteeism refers to the phenomenon whereby employees continue to attend work while unwell. Existing research suggests that presentee workers may suffer consequences to their health and mental strain. In this paper, we investigate whether such consequences also have downstream effects in terms of the errors people make at work. We studied the effects of presenteeism among a large sample of pharmacists (N = 1,205), an occupation in which errors made can be safety critical, with implications for patient health. Seventy-six percent of the pharmacists in our sample were classed as presentee, having attended work while unwell enough to have taken time off on at least two occasions over the previous year. Presentee pharmacists made significantly more minor errors and serious mistakes, such as dispensing errors, compared to nonpresentee pharmacists. They also experienced greater feelings of anxiety and depression. Mediation analyses suggested that higher anxiety rates explained why presentee employees made more errors at work. Presenteeism therefore has significant health costs for both workers and their beneficiaries and can be classed as an important work-related stressor.

Full text

The Hidden Dangers of Attending Work While
Unwell: A Survey Study of Presenteeism
Among Pharmacists
Karen Niven
University of Manchester
Natalia Ciborowska
Randstad Professionals, Pozna´
n, Poland
Presenteeism refers to the phenomenon whereby employees continue to attend
work while unwell. Existing research suggests that presentee workers may suffer
consequences to their health and mental strain. In this paper, we investigate
whether such consequences also have downstream effects in terms of the errors
people make at work. We studied the effects of presenteeism among a large
sample of pharmacists (N ⫽1,205), an occupation in which errors made can be
safety critical, with implications for patient health. Seventy-six percent of the
pharmacists in our sample were classed as presentee, having attended work
while unwell enough to have taken time off on at least two occasions over the
previous year. Presentee pharmacists made significantly more minor errors and
serious mistakes, such as dispensing errors, compared to nonpresentee pharma-
cists. They also experienced greater feelings of anxiety and depression. Media-
tion analyses suggested that higher anxiety rates explained why presentee
employees made more errors at work. Presenteeism therefore has significant
health costs for both workers and their beneficiaries and can be classed as an
important work-related stressor.
Keywords: presenteeism, errors, anxiety, depression, pharmacists
Illness is an inevitable part of the human condition. Taking time off work
due to illness (i.e., absenteeism) may be inconvenient to both workers and their
employing organizations, due to issues including increased workload on return to
work and difficulties and costs associated with finding replacements. For such
Karen Niven, Manchester Business School, University of Manchester; Natalia Ci-
borowska, Randstad Professionals, Pozna´
n, Poland.
Correspondence concerning this article should be addressed to Karen Niven, Manchester
Business School, University of Manchester, Manchester M15 6PB, England. E-mail: karen
.niven@mbs.ac.uk
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
International Journal of Stress Management © 2015 American Psychological Association
2015, Vol. 22, No. 2, 207–221 1072-5245/15/$12.00 http://dx.doi.org/10.1037/a0039131
207

reasons, a trend termed “presenteeism” is emerging in organizations, whereby
people continue to attend work even when they are ill. Although at first glance
presenteeism appears to address some of the issues associated with absenteeism,
it may lead to other problems. Research suggests that attending work while
unwell can reduce workers’ productivity, worsen the health of presentee workers,
and lead to the spread of illness and may therefore have even greater costs than
absenteeism (Goetzel et al., 2004).
In spite of mounting evidence for its importance, presenteeism has tradi-
tionally been a much underresearched area. According to Dew and colleagues, as
of a July 2003 search, the number of published research articles concerning
presenteeism was a mere .01% of those regarding absenteeism (Dew, Keefe, &
Small, 2005). Thus, there are significant gaps within our knowledge of presen-
teeism. In particular, while presenteeism has been linked to negative conse-
quences for employees, such as poorer mental and physical health, it is unclear
whether employees who attend work while unwell are more prone to making
errors. Errors can have important consequences, particularly for those working in
safety-critical occupations. Yet while stressors that cause workers’ health to
suffer have been implicated more generally as key predictors of worker errors
(Vidyarthi, Auerbach, Wachter, & Katz, 2007), the link between presenteeism
and making errors is currently poorly understood. Moreover, there are many
occupational groups for whom the consequences of presenteeism have yet to be
identified. For example, while presenteeism is known to be most prevalent
among health care professionals (Aronsson, Gustafsson, & Dallner, 2000), little,
if any, research has focused on pharmacists. Not only might pharmacists expe-
rience negative consequences for their own health if they attend work while
unwell, but there may also be grave implications for their patients, as an increase
in employee errors could potentially lead to fatal consequences (Phipps, Noyce,
Parker, & Ashcroft, 2009).
The current paper presents a large-scale survey study that extends the
literature on presenteeism in two key ways. First, we provide a test of whether
presenteeism is associated with higher work-related error rates. In doing so, we
explore the effect of presenteeism on mental strain as a potential mechanism for
the effects on error rates. Second, we examine presenteeism among pharmacists,
providing the first evidence (to our knowledge) of the prevalence and effects of
presenteeism in this underresearched occupational group.
PRESENTEEISM
There is some debate in the literature regarding the definition of presen-
teeism. Early accounts used the term to refer to individuals who exhibited
excellent attendance at work (Canfield & Soash, 1955) or to those who
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
208 Niven and Ciborowska

simply attended work as opposed to being absent (Smith, 1970). Others have
used the term presenteeism to refer to the tendency to be reluctant to work
part time rather than full time (Sheridan, 2004) or have equated presenteeism
with a kind of “survivor syndrome” whereby people stay at work beyond the
time needed to perform effectively (Simpson, 1998). Nevertheless, in recent
years researchers have started to converge on a common definition of
presenteeism as referring to situations wherein workers continue to attend
work even though they feel unwell enough to take sickness absence (e.g.,
Aronsson & Gustafsson, 2005;Aronsson et al., 2000;Johns, 2011). In
accordance with this definition, estimates of the prevalence of presenteeism
usually classify someone as presentee if they have attended work while
unwell at least twice over a yearlong period. Such estimates have ranged
between 53% and 72% in recent population studies (Aronsson & Gustafsson,
2005;Caverley, Cunningham, & MacGregor, 2007).
The trend toward presenteeism that has emerged in organizations is
troubling from a health perspective, as people may find it more difficult to
recover from illness and may develop complications or pass their illness on
to others (Broadhead, Blazer, George, & Tse, 1990;Kivimäki et al., 2005).
For example, Rosvold and Bjertness (2001) reported that over a 1-year
period, more than half of the Norwegian physicians they surveyed had
attended work while having an infectious disease that could be transmitted to
their patients.
However, recent conceptualizations of presenteeism advocate taking a
psychological, rather than medical, approach to understanding the phenom-
enon and the dangers it presents (Johns, 2011). Presenteeism is thought to be
strongly interlinked with stress at work, and studies of presenteeism have
reported that job stress is an important trigger of staying at work even though
one is unwell. For example, Elstad and Vabø’s (2008) survey study of
Scandinavian care workers found that as job stress increased, the level of
sickness presenteeism rose sharply and to a greater extent than did the level
of absenteeism. A further study of Sweden council employees reported that
rates of sickness presence were significantly higher among employees who
suffered burnout and emotional exhaustion compared with a nonburnout
group (Peterson et al., 2008). Longitudinal research has further explicated
that job stress and presenteeism are causally linked in both directions, with
presenteeism also leading to burnout over time (Demerouti, Le Blanc, Bak-
ker, Schaufeli, & Hox, 2009).
Not only are employees who go to work when feeling ill more prone to
experiencing signs of strain and mental ill health (e.g., anxiety, depression;
Bergström et al., 2009), but these negative consequences are also thought to
compromise workers’ ability to do their job (Burton, Conti, Chen, Schultz, &
Edington, 2002;Lerner et al., 2004). For such reasons, researchers increas-
ingly suggest that going to work while unwell may have more significant
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
209Effects of Presenteeism on Safety-Critical Errors

costs than taking time off. A review integrating the results of five large
studies reported that costs of presenteeism (operationalized as on-the-job
productivity losses) associated with the 10 most costly health conditions were
higher than the costs of absenteeism resulting from these same conditions
(Goetzel et al., 2004). For headache/migraine, allergies, arthritis, asthma, and
mental illness, presenteeism costs accounted for over 70% of the overall costs
associated with the condition.
Presenteeism and Work Errors
One of the potential costs of presenteeism that has to date been unex-
plored concerns its links with workplace errors. Errors are an unfortunate
reality within all industrial sectors and can have devastating consequences for
employees and their organizations, especially in safety-critical occupations.
For example, within health care contexts, medication errors are a leading
cause of unintended harm to patients (Fogarty & McKeon, 2006).
Most theories that seek to explain how and why errors occur in organi-
zations suggest a contributory role of work-related stressors. For example,
Hockey’s (1997) theory explains how stressors such as high workload
influence performance and errors. The idea that underlies Hockey’s theory is
that compensatory mechanisms typically kick in under conditions of stress,
such that people seek to protect their performance by recruiting and allocat-
ing further resources. However, because people’s resources are limited, this
process comes at the expense of mental strain, which in turn can produce
behavioral costs. In other words, when workers are under high pressure, they
are only able to maintain normal levels of performance by risking their
psychological health, making the probability of errors substantially larger.
Presenteeism can be conceptualized according to this theory as a form of
stressor that the worker chooses, feels obligated, or is obligated to engage in.
When physically unwell and continuing to attend work, maintenance of
ordinary levels of performance will require greater compensatory effort (e.g.,
higher concentration, battling to overcome symptoms that may adversely
affect work) on the part of the employee, which is likely to lead to mental
strain. In support of this notion, several studies have reported strong causal
links between presenteeism and mental strain (e.g., Bergström et al., 2009;
Demerouti et al., 2009). In turn, this greater level of strain is likely to lead to
higher rates of errors, as the capacity for paying attention to environmental
stimuli and monitoring one’s behavior is compromised. Indeed, mental strain
has been implicated as a major cause of workplace errors (e.g., Fogarty &
McKeon, 2006;Holden et al., 2010;Vidyarthi et al., 2007).
The expected link between presenteeism and errors is particularly troubling
within health care occupations, not just because of the importance of errors in
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
210 Niven and Ciborowska

such contexts, but also because of the prevalence of presenteeism. Presenteeism
rates among health care professionals are among the highest of any occupational
group (Aronsson et al., 2000;Elstad & Vabø, 2008). For example, despite
guidelines clearly stating that doctors have an ethical duty to ensure that their
own health problems are effectively managed, the majority of United Kingdom
(UK)–based consultants and general practitioners surveyed indicated that they
would not take time off work when experiencing a range of health problems,
including vomiting all night, unexplained headaches, and a growing dependence
on alcohol (Forsythe, Calnan, & Wall, 1999). Commonly cited reasons for the
high prevalence of presenteeism among health care workers include viewing
daily tasks as “must-do” tasks and perceiving a difficulty in finding replacements
(e.g., Aronsson et al., 2000).
Although researchers have intensively studied presenteeism among many
health care occupations (e.g., hospital physicians, medical residents, general
practitioners, nurses), pharmacists have so far been relatively neglected. This
is despite the fact that the factors that typically lead to high rates of
presenteeism in other health care occupations are likely to be equally present
among pharmacists and that pharmacists work within an environment in
which errors (e.g., dispensing errors) can compromise patient health and, in
the most extreme cases, may cost a patient’s life (Phipps et al., 2009).
In the present study, we build on existing research by studying associations
between presenteeism and error rates among pharmacists. In line with the
theoretical and empirical evidence discussed above, we expect that higher levels
of mental strain (e.g., anxiety and depression) will account for the links between
presenteeism and errors. To establish the relative importance of presenteeism as
a contributing factor to work-related strain and errors, we investigate the effects
of presenteeism over and above the effects of absenteeism.
Hypothesis 1: Presentee employees will report higher levels of errors
than nonpresentee employees.
Hypothesis 2: The link between presenteeism and error rates will be
mediated by mental strain.
METHOD
Participants
Participants were members of the Pharmacists’ Defense Association
(PDA), a not-for-profit organization that represents the interests of pharma-
cists in the UK. In the UK, there are no nationwide sick leave policies for
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
211Effects of Presenteeism on Safety-Critical Errors

pharmacists, meaning that policies vary across employing organizations. An
e-mail was sent out to all registered PDA members that advertised the study
as an investigation of pharmacists’ work experiences and that included a link
to an online survey. At the time, around 19,000 pharmacists were members
of the PDA. The survey was available for a 1-month period, and 1,205
completed responses were returned (representing an approximate response
rate of 6.3%). The sample (63% females, mean age ⫽41.12 years, SD ⫽
12.70) had an average of 17.05 years of experience (SD ⫽13.06) working in
a pharmacy. Seventy-seven percent of the sample worked in community
pharmacies; 17% worked in hospital pharmacies; and 6% worked in Primary
Care Trusts (PCTs, which are UK National Health Service administrative
bodies responsible for commissioning primary, community, and secondary
health services from providers). Of those working in community pharmacies,
12% worked in single shops, while the remaining 88% worked for pharmacy
chains.
Measures
The survey comprised a series of self-reported measures. The measures
were presented in the following order: control variables, presenteeism, ab-
senteeism, errors, and mental strain.
Presenteeism and Absenteeism
Respondents were asked to complete established measures of presentee-
ism and absenteeism developed by Aronsson et al. (2000) and validated in
numerous research studies (e.g., Aronsson & Gustafsson, 2005;Hansen &
Andersen, 2008). The measures asked: “How many times during the last 12
months have you taken sick leave outside of your annual leave?” (for
absenteeism) and “How many times during the last 12 months have you gone
to work even though it would have been reasonable to take sick leave?” (for
presenteeism). We extended the original 4-point response scale to include
seven options: not relevant, have not been sick over the previous 12 months
(1), none (2), once (3), 2–3 times (4), 4–5 times (5), 6 –10 times (6), and more
than 10 times (7). The measures were developed to be used as dichotomous,
with the cutoff point being more than or less than two instances of presen-
teeism (or absenteeism) across the year, to match the accepted definition of
presenteeism (Aronsson et al., 2000), and most studies in the literature retain
this scoring system. In keeping with this, in the present study we dichoto-
mized the items for analysis, with Responses 1–3 coded as “0” and 4–7
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
212 Niven and Ciborowska

(representing two or more incidents of absence or presence over the last year)
coded as “1.”
Errors
The frequency with which errors were made by the pharmacists was
assessed using two self-reported items. Although self-reported data on errors
could be subject to some biases (as discussed later in the paper), self-reported
measures of errors are common in the literature in health care (e.g., Fogarty
& McKeon, 2006;Vidyarthi et al., 2007) and non-health care (e.g., Kecklund
& Svenson, 1997) settings due to the difficulty in obtaining accurate objec-
tive data from organizations. The items regarded minor and serious errors,
respectively, and were developed in consultation with a focus group of
pharmacists concerning the types of errors they (or others they knew) had
made at work. The pharmacists identified that medication errors with poten-
tially severe consequences were more serious than other, more minor (i.e.,
less consequential) errors that could be made (e.g., an error that led to no
patient harm, such as giving incorrect informal advice) and so suggested that
they might be meaningfully distinguished. After developing the two items,
face validity was checked with another, independent sample of pharmacists.
Both items asked about errors made over the previous 4 weeks in order
to avoid retrospective biases that can affect recall of specific incidents over
longer time periods (Lawrence, Roy, Harikrishnan, Yu, & Dabbous, 2013).
Specifically, respondents were asked: “How many times within the last 4
weeks have you made a minor work-related mistake when at work?” and
“How many times within the last 4 weeks have you made a more serious
work-related mistake when at work (e.g., giving a patient the wrong medi-
cation, prescribing the wrong dosage of a medication, or prescribing medi-
cation of the wrong strength)?” We chose to specify the nature of serious
errors to signal that these would be errors relating to medication with
potentially severe consequences. In contrast, the nature of minor errors was
not specified due to the diverse range of examples offered within our focus
group. Response options were none (1), none (2), 2–3 times (3), 4 –5 times
(4), 6 –10 times (5), and more than 10 times (6).
Mental Strain
Mental strain was measured using established scales developed by Warr
(1990), assessing the extent to which pharmacists had experienced feelings of
anxiety (␣⫽.92) and depression (␣⫽.94) while at work over the past month.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
213Effects of Presenteeism on Safety-Critical Errors

Control Variables
Sex, age, and years of working experience were measured as control
variables for the present study. These variables were selected because they
are typically controlled for in studies of presenteeism, mental strain, and
errors. Although findings vary from study to study, there is some indication
that presenteeism and mental strain might be more prevalent among females
than males and that age and tenure may be positively associated with these
outcomes (e.g., Aronsson et al., 2000;Sevastos, Smith, & Cordery, 1992).
Data Analysis
Data were analyzed using SPSS version 20. Hypothesis 1 was analyzed
using multiple analysis of covariance (MANCOVA), in which the differences
between presentee and nonpresentee pharmacists were examined simultane-
ously on four outcome variables: minor errors, serious errors, anxiety, and
depression. Absenteeism was included as a second factor in the model to
identify the relative risks of pharmacists staying at work versus taking
time off work when unwell. Hypothesis 2, which predicted mediation of
the effect of presenteeism on errors via mental strain, was tested in two
ways. First, the procedures of Baron and Kenny (1986) were followed, in
which the effect of presenteeism on error rates was tested in the presence
of the additional predictors of anxiety and depression in a regression
analysis. Second, the procedures of Hayes (2009) were followed, with the
significance of the indirect effects of presenteeism on error rates via
anxiety and depression tested simultaneously using bootstrapping proce-
dures. Bootstrapping is recommended over alternative procedures (e.g.,
the Sobel test) because it avoids unrealistic assumptions about the shape
of the sampling distribution of indirect effects. Instead, 1,000 resamples
are drawn to estimate 95% bias-corrected bootstrap confidence intervals
(CIs) for all indirect effects. If the CIs do not include 0, the indirect effect
is said to be significant. In all analyses, the three control variables were
included (sex, age, and years of experience).
RESULTS
Of the overall sample, 76% was classed as presentee, having attended
work while unwell enough to have taken time off at least twice in the past
year. Rates of presenteeism differed slightly across pharmacy types, with
rates highest in hospital pharmacies (83%) compared to community pharma-
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
214 Niven and Ciborowska

cies (75%) and PCTs (70%), ␹
2
(2) ⫽6.13, p⬍.05. A smaller proportion of
the sample (43%) was classed as absentee, having taken time off while
unwell at least twice in the past year. Again, absenteeism rates were highest
in hospital pharmacies (65%) compared to community pharmacies (38%) and
PCTs (47%), ␹
2
(2) ⫽47.60, p⬍.01.
The results of the MANCOVA revealed a significant multivariate effect
of presenteeism, F(4, 1,117) ⫽21.26, p⬍.01, on the outcome variables.
Univariate analyses confirmed that there were significant differences between
presentee and nonpresentee employees in all four outcomes (see Table 1). In
support of Hypothesis 1, presentee employees made more minor and serious
errors compared with nonpresentee employees. They also experienced higher
levels of anxiety and depression.
No multivariate effect of absenteeism was observed in the MANCOVA,
F(4, 1,117) ⫽1.01, p⫽.40. Similarly, there was no significant interaction
between absenteeism and presenteeism, F(4, 1,117) ⫽2.07, p⫽.08. These
findings suggest that being unwell in and of itself did not affect pharmacists’
mental strain and errors made, because taking sick leave did not produce any
additional detriments above and beyond remaining at work while unwell. Of
the control variables, sex, F(4, 1,117) ⫽6.43, p⬍.01, and years of
experience, F(4, 1,117) ⫽2.50, p⬍.05, both had significant multivariate
effects. Univariate tests indicated that sex predicted serious error rates, F(1,
1,112) ⫽14.96, p⬍.01, with more errors reported by males. Years of
experience predicted pharmacists’ anxiety, F(1, 1,112) ⫽8.68, p⬍.01, with
higher years of experience associated with lower anxiety.
Mediation analysis was then run to test whether mental strain was
responsible for the effect of presenteeism on pharmacists’ errors. The results
of regression (in Table 2) show support for this proposition; presenteeism
was no longer a significant predictor of either minor (Model 1) or serious
(Model 2) errors when mental strain was controlled for, while anxiety (but
not depression) did have a significant effect on both outcomes. Moreover, the
Table 1. Differences Between Presentee and Nonpresentee Employees in Core
Study Outcomes
Presentee
employees
(N⫽918)
Nonpresentee
employees
(N⫽287)
Difference F(df)Mean SD Mean SD
Minor errors 2.47 .04 2.20 .08 8.83 (1, 1,112)
**
Serious errors 1.38 .02 1.26 .05 4.80 (1, 1,112)
*
Anxiety 3.57 .03 2.91 .07 80.26 (1, 1,112)
**
Depression 3.00 .04 2.34 .08 57.35 (1, 1,112)
**
Note. Sex, age, and years of experience were controlled for in these analyses.
*
p⬍.05.
**
p⬍.01.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
215Effects of Presenteeism on Safety-Critical Errors

results of bootstrapping to test the significance of the indirect effects con-
firmed a significant indirect effect of presenteeism via anxiety for minor
errors (estimate ⫽.16, 95% CI [.10, .24]) and serious errors (estimate ⫽.07,
95% CI [.04, .12]). In contrast, the indirect effects of presenteeism via
depression were not significant (minor errors, estimate ⫽⫺.02, 95% CI
[⫺.07, .03]; serious errors, estimate ⫽.00, 95% CI [⫺.03, .04]). Hypothesis
2 was therefore partially supported; the feelings of anxiety that arise from
presenteeism explain why people who attend work while unwell make more
errors in their work.
DISCUSSION
The current study presents evidence that presenteeism affects the rates
with which pharmacists make both minor errors and more serious errors, such
as giving a patient the wrong medication or prescribing the wrong dosage of
a medication. The effect of presenteeism was observed over and above the
effect of absenteeism. Thus, increased error rates were not simply a product
of illness; they were specifically a product of being unwell and continuing to
attend work. Those pharmacists who remained at work while unwell also
experienced higher feelings of anxiety and depression, with the higher
anxiety shown to be responsible for the increased error rates.
The effects observed are consistent with Hockey’s (1997) theory con-
cerning how stressors affect work performance. While working when unwell,
the pharmacists would have needed to exert compensatory effort to maintain
ordinary performance levels, which would lead to higher anxiety and depres-
sion. In turn, this mental strain would make the likelihood of errors greater.
The finding that anxiety rather than depression was responsible for higher
errors is consistent with the literature on well-being, which conceptualizes
anxiety and depression as unpleasant states that differ primarily based on
their level of arousal (e.g., Warr, 1990). As the high arousal state, anxiety
Table 2. Mediated Effects of Presenteeism on Error Rates Via Mental Strain
Model 1 Model 2
Effect on minor errors Effect on serious errors
Predictors ␤SE t ␤SE t
Sex ⫺.15 .06 ⫺2.40
*
⫺.15 .04 ⫺3.86
**
Age .01 .01 0.75 ⬍.01 .01 0.58
Experience ⫺.01 .01 ⫺0.75 ⬍.01 .01 ⫺0.82
Presenteeism .10 .08 1.29 ⬍.01 .05 0.05
Anxiety .27 .05 5.78
**
.12 .03 4.32
**
Depression ⫺.03 .04 0.75 .01 .02 0.21
*
p⬍.05.
**
p⬍.01.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
216 Niven and Ciborowska

could be seen as more likely to interfere with the types of cognitive processes
(e.g., attention, monitoring) that are necessary to avoid errors (Tobias, 1985).
Depression, in contrast, is low arousal and so may not interfere as much with
these processes.
Our findings of 76% prevalence of presenteeism among this population
are slightly higher than the prevalence rates reported in previous studies in
the working population, with estimates ranging between 53% and 72%
(Aronsson & Gustafsson, 2005;Caverley et al., 2007). However, this is in
line with accounts that report prevalence as particularly high among health
care workers (e.g., Aronsson et al., 2000), who appear to take their own
health concerns less seriously than those of their patients (Rosvold & Bjert-
ness, 2001). Moreover, the findings largely concur with the growing body of
evidence suggesting that presenteeism may be more costly for organizations
than absenteeism (Goetzel et al., 2004). In the present case, presenteeism, but
not absenteeism, was linked to the outcomes of anxiety, depression, and
errors, all of which could lead to costs such as reduced productivity, further
periods of illness, and litigation.
To the best of our knowledge, this is the first study to demonstrate that
not only might attending work while unwell have implications for workers’
own health, it might also lead to significant errors that could compromise
other people’s health. In addition, the study is the first to explore the
prevalence and effects of presenteeism among pharmacists. Although pre-
senteeism has been investigated in other health care professions, such as
physicians and nurses (e.g., Aronsson et al., 2000;Forsythe et al., 1999),
pharmacists have to date been relatively neglected. The study involved a
large sample, with representation from different types of pharmacists (in-
cluding those working in hospitals, PCTs, and community pharmacies).
Nevertheless, the study does include some weaknesses that should be
recognized. In particular, the cross-sectional study design means that our
findings show a link between presenteeism, anxiety, and error rates but do not
prove a causal order. Although prospective studies are relatively rare in this
area, those reported support the causal order proposed here; for example,
Bergström and colleagues (2009) reported a causal link between presentee-
ism and mental ill health at 18 months and 3 years. The use of self-reported
data in the present study could also have led to socially desirable patterns of
responding. However, as Johns (2011) argues, “it is hard to conceive of a
measure of presenteeism that would not use self-report” (p. 496), and the high
reported incidence suggests that social desirability was unlikely to be a
problem in terms of the reporting of presenteeism in this study. Moreover,
although it could be argued that pharmacists might have underreported errors
to portray themselves positively, we do not believe that this issue would have
strongly compromised our findings, in that workers are likely to be more
inclined to respond truthfully to an anonymous, independent survey than to
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
217Effects of Presenteeism on Safety-Critical Errors

report errors in official systems to their organizations, which may go on their
employment records and be used in official decisions (e.g., about promo-
tions). A related issue, however, is that the order of question presentation in
the survey could have affected responses (e.g., participants who reported
making errors might have been more likely to report suffering mental strain
as a possible justification for those errors).
A further limitation of the research is the low response rate. While the
sample in the study is certainly large, only 6.3% of the members of the PDA
at the time actually completed the survey. The low response may be due to
issues in contacting members, many of whom may not have a current e-mail
address registered with the PDA or may not regularly check their e-mail; the
lack of direct incentives for participation; or the overworked nature of the
target sample (Phipps et al., 2009). Advertising the study as an investigation
of pharmacists’ work experiences helped to mitigate issues of bias in the
sample in terms of having a vested interest in the issue of presenteeism, but
it is possible that the final sample is not completely representative of the
membership of the PDA and thus of UK pharmacists as a whole. Finally,
although we controlled for some of the demographic factors that may be
related to presenteeism in our analysis (i.e., age, sex, years of experience), we
did not account for other demographic factors, such as ethnicity, marital
status, and number of dependents, which may also be relevant factors in
influencing workers’ sense of responsibility and ability to be present at work.
Nor did we consider working time arrangements (Bockerman & Laukkanen,
2010), worker attitudes, and health status.
Future studies on the link between presenteeism and errors should take
some of the factors outlined above into account to determine whether
presenteeism is still associated with error rates over and above these possible
confounds. A qualitative or mixed-methods approach might also be useful to
obtain richer data concerning the actual errors that are made under situations
of presenteeism. In turn, this might facilitate a more comprehensive estima-
tion of the costs associated with presenteeism in specific industry settings,
such as pharmacies.
Future research should also investigate potential differences associated
with presenteeism caused by different types of illnesses to determine whether
anxiety and therefore error rates vary by condition. There are many people
who have chronic health conditions (e.g., lower back pain) who need to
continue to work rather than taking absences every time the problem arises.
It will be important for future research to establish whether presenteeism in
such conditions poses less threat to the worker (and potentially less threat to
others) and what types of measures can be taken by the organization to
prevent presenteeism in these conditions becoming detrimental. Further
investigation of the contextual factors that contribute to presenteeism, such as
job insecurity, managerial pressure, or difficulty in finding a replacement,
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
218 Niven and Ciborowska

will also be important in informing interventions to reduce presenteeism for
those who have acute health conditions.
The emerging trend toward staying at work when unwell may offset
some of the more immediate costs associated with absenteeism, such as
replacing staff. However, the present study suggests that this trend could have
devastating consequences for the well-being of pharmacists and their pa-
tients. Reducing presenteeism and its associated concerns should therefore be
a high priority for policy makers. Our findings suggest that there is a need to
monitor the health of employees who are attending work and to establish
whether there is any external pressure (e.g., from managers) or internal
pressure (e.g., fear of losing one’s job) to attend work even when unwell. Any
organizational practices that put undue pressure on employees to be present
when unwell (e.g., policies whereby sick leave is prohibited unless a replace-
ment can be found) should be strongly discouraged. In terms of actually
reducing presenteeism, Wrate (1999) opines that in the case of doctors at
least, “only a substantial increase in medical staffing may noticeably reduce
doctors’ presenteeism” (p. 1502). An alternative or at least complementary
approach suggested by the present findings would be to tackle the conse-
quences of presenteeism. In the present study, the main reason why errors
occurred through presenteeism was the increased anxiety experienced by
presentee workers. Increasing job autonomy (Wieclaw et al., 2008) or using
psychoeducational interventions (Donker, Griffiths, Cuijpers, & Christensen,
2009) could help reduce anxiety among workers, which in turn could reduce
error rates.
REFERENCES
Aronsson, G., & Gustafsson, K. (2005). Sickness presenteeism: Prevalence, attendance-
pressure factors, and an outline of a model for research. Journal of Occupational and
Environmental Medicine, 47, 958–966. http://dx.doi.org/10.1097/01.jom.0000177219
.75677.17
Aronsson, G., Gustafsson, K., & Dallner, M. (2000). Sick but yet at work: An empirical study
of sickness presenteeism. Journal of Epidemiology and Community Health, 54, 502–509.
http://dx.doi.org/10.1136/jech.54.7.502
Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social
psychological research: Conceptual, strategic, and statistical considerations. Journal of
Personality and Social Psychology, 51, 1173–1182. http://dx.doi.org/10.1037/0022-3514
.51.6.1173
Bergström, G., Bodin, L., Hagberg, J., Lindh, T., Aronsson, G., & Josephson, M. (2009). Does
sickness presenteeism have an impact on future general health? International Archives of
Occupational and Environmental Health, 82, 1179–1190. http://dx.doi.org/10.1007/
s00420-009-0433-6
Böckerman, P., & Laukkanen, E. (2010). What makes you work while you are sick? Evidence
from a survey of workers. European Journal of Public Health, 20, 43–46. http://dx.doi
.org/10.1093/eurpub/ckp076
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
219Effects of Presenteeism on Safety-Critical Errors

Broadhead, W. E., Blazer, D. G., George, L. K., & Tse, C. K. (1990). Depression, disability
days, and days lost from work in a prospective epidemiologic survey. Journal of the
American Medical Association, 264, 2524–2528. http://dx.doi.org/10.1001/jama.1990
.03450190056028
Burton, W. N., Conti, D. J., Chen, C. Y., Schultz, A. B., & Edington, D. W. (2002). The
economic burden of lost productivity due to migraine headache: A specific worksite
analysis. Journal of Occupational and Environmental Medicine, 44, 523–529. http://dx
.doi.org/10.1097/00043764-200206000-00013
Canfield, G. W., & Soash, D. G. (1955). Presenteeism—A constructive view. Industrial
Medicine and Surgery, 24, 417–418.
Caverley, N., Cunningham, J. B., & MacGregor, J. N. (2007). Sickness presenteeism, sickness
absenteeism, and health following restructuring in a public service organization. Journal
of Management Studies, 44, 304–319. http://dx.doi.org/10.1111/j.1467-6486.2007
.00690.x
Demerouti, E., Le Blanc, P. M., Bakker, A. B., Schaufeli, W. B., & Hox, J. (2009). Present but
sick: A three-wave study on job demands, presenteeism and burnout. Career Development
International, 14, 50– 68. http://dx.doi.org/10.1108/13620430910933574
Dew, K., Keefe, V., & Small, K. (2005). “Choosing” to work when sick: Workplace presen-
teeism. Social Science & Medicine, 60, 2273–2282. http://dx.doi.org/10.1016/j.socscimed
.2004.10.022
Donker, T., Griffiths, K. M., Cuijpers, P., & Christensen, H. (2009). Psychoeducation for
depression, anxiety and psychological distress: A meta-analysis. BMC Medicine, 7, 79.
http://dx.doi.org/10.1186/1741-7015-7-79
Elstad, J. I., & Vabø, M. (2008). Job stress, sickness absence and sickness presenteeism in
Nordic elderly care. Scandinavian Journal of Public Health, 36, 467–474. http://dx.doi
.org/10.1177/1403494808089557
Fogarty, G. J., & McKeon, C. M. (2006). Patient safety during medication administration: The
influence of organizational and individual variables on unsafe work practices and medi-
cation errors. Ergonomics, 49 (5–6), 444 – 456. http://dx.doi.org/10.1080/
00140130600568410
Forsythe, M., Calnan, M., & Wall, B. (1999). Doctors as patients: Postal survey examining
consultants and general practitioners adherence to guidelines. British Medical Journal,
319, 605–608. http://dx.doi.org/10.1136/bmj.319.7210.605
Goetzel, R. Z., Long, S. R., Ozminkowski, R. J., Hawkins, K., Wang, S., & Lynch, W. (2004).
Health, absence, disability, and presenteeism cost estimates of certain physical and mental
health conditions affecting U.S. employers. Journal of Occupational and Environmental
Medicine, 46, 398– 412. http://dx.doi.org/10.1097/01.jom.0000121151.40413.bd
Hansen, C. D., & Andersen, J. H. (2008). Going ill to work—What personal circumstances,
attitudes and work-related factors are associated with sickness presenteeism? Social
Science & Medicine, 67, 956–964. http://dx.doi.org/10.1016/j.socscimed.2008.05.022
Hayes, A. F. (2009). Beyond Baron and Kenny: Statistical mediation analysis in the new
millennium. Communication Monographs, 76, 408– 420. http://dx.doi.org/10.1080/
03637750903310360
Hockey, G. R. J. (1997). Compensatory control in the regulation of human performance under
stress and high workload; a cognitive-energetical framework. Biological Psychology,
45 (1–3), 73–93. http://dx.doi.org/10.1016/S0301-0511(96)05223-4
Holden, R. J., Patel, N. R., Scanlon, M. C., Shalaby, T. M., Arnold, J. M., & Karsh, B. T.
(2010). Effects of mental demands during dispensing on perceived medication safety and
employee well-being: A study of workload in pediatric hospital pharmacies. Research in
Social & Administrative Pharmacy, 6, 293–306. http://dx.doi.org/10.1016/j.sapharm.2009
.10.001
Johns, G. (2011). Attendance dynamics at work: The antecedents and correlates of presentee-
ism, absenteeism, and productivity loss. Journal of Occupational Health Psychology, 16,
483–500. http://dx.doi.org/10.1037/a0025153
Kecklund, L. J., & Svenson, O. (1997). Human errors and work performance in a nuclear power
plant control room: Associations with work-related factors and behavioral coping. Reli-
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
220 Niven and Ciborowska

ability Engineering & System Safety, 56, 5–15. http://dx.doi.org/10.1016/S0951-
8320(96)00137-8
Kivimäki, M., Head, J., Ferrie, J. E., Hemingway, H., Shipley, M. J., Vahtera, J., & Marmot,
M. G. (2005). Working while ill as a risk factor for serious coronary events: The Whitehall
II study. American Journal of Public Health, 95, 98–102. http://dx.doi.org/10.2105/AJPH
.2003.035873
Lawrence, C., Roy, A., Harikrishnan, V., Yu, S., & Dabbous, O. (2013). Association between
severity of depression and self-perceived cognitive difficulties among full-time employ-
ees. The Primary Care Companion for CNS Disorders, 15(3), pii: PCC.12m01469.
http://dx.doi.org/10.4088/PCC.12m01469
Lerner, D., Adler, D. A., Chang, H., Lapitsky, L., Hood, M. Y., Perissinotto, C.,...Rogers, W. H. (2004).
Unemployment, job retention, and productivity loss among employees with depression. Psychiatric
Services, 55, 1371–1378. http://dx.doi.org/10.1176/appi.ps.55.12.1371
Peterson, U., Demerouti, E., Bergström, G., Samuelsson, M., Åsberg, M., & Nygren, A. (2008).
Burnout and physical and mental health among Swedish healthcare workers. Journal of
Advanced Nursing, 62, 84–95. http://dx.doi.org/10.1111/j.1365-2648.2007.04580.x
Phipps, D. L., Noyce, P. R., Parker, D., & Ashcroft, D. M. (2009). Medication safety in
community pharmacy: A qualitative study of the sociotechnical context. BMC Health
Services Research, 9, 158–168. http://dx.doi.org/10.1186/1472-6963-9-158
Rosvold, E. O., & Bjertness, E. (2001). Physicians who do not take sick leave: Hazardous
heroes? Scandinavian Journal of Public Health, 29, 71–75. http://dx.doi.org/10.1177/
14034948010290010101
Sevastos, P., Smith, L., & Cordery, J. L. (1992). Evidence on the reliability and construct
validity of Warr’s (1990) well-being and mental health measures. Journal of Occupational
and Organizational Psychology, 65, 33–49. http://dx.doi.org/10.1111/j.2044-8325.1992
.tb00482.x
Sheridan, A. (2004). Chronic presenteeism: The multiple dimensions to men’s absence from
part-time work. Gender, Work and Organization, 11, 207–225. http://dx.doi.org/10.1111/
j.1468-0432.2004.00229.x
Simpson, R. (1998). Presenteeism, power and organizational change: Long hours as a career
barrier and the impact on the working lives of women managers. British Journal of
Management, 9, 37–50. http://dx.doi.org/10.1111/1467-8551.9.s1.5
Smith, D. J. (1970). Absenteeism and “presenteeism” in industry. Archives of Environmental
Health, 21, 670– 677. http://dx.doi.org/10.1080/00039896.1970.10667313
Tobias, S. (1985). Test anxiety: Interference, defective skills, and cognitive capacity. Educa-
tional Psychologist, 20, 135–142. http://dx.doi.org/10.1207/s15326985ep2003_3
Vidyarthi, A. R., Auerbach, A. D., Wachter, R. M., & Katz, P. P. (2007). The impact of duty
hours on resident self reports of errors. Journal of General Internal Medicine, 22,
205–209. http://dx.doi.org/10.1007/s11606-006-0065-4
Warr, P. (1990). The measurement of well-being and other aspects of mental health. Journal of
Occupational Psychology, 63, 193–210. http://dx.doi.org/10.1111/j.2044-8325.1990
.tb00521.x
Wieclaw, J., Agerbo, E., Mortensen, P. B., Burr, H., Tuchsen, F., & Bonde, J. P. (2008). Psychosocial
working conditions and the risk of depression and anxiety disorders in the Danish workforce. BMC
Public Health, 8, 280. http://dx.doi.org/10.1186/1471-2458-8-280
Wrate, R. M. (1999). Increase in staff numbers may reduce doctors’ “presenteeism.” BMJ, 319,
1502. http://dx.doi.org/10.1136/bmj.319.7223.1502a
Received June 13, 2014
Revision received February 19, 2015
Accepted February 23, 2015 䡲
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
221Effects of Presenteeism on Safety-Critical Errors