ChapterPDF Available

Accident under-reporting in the workplace

Accident Underreporting 1
Accident Underreporting in the Workplace
Tahira M. Probst Erica L. Bettac Christopher Austin
Washington State University Vancouver
Probst, T. M., Bettac, E., & Austin, C. (2019). Accident underreporting in the workplace. In R. Burke & A.
Richardsen (Eds.), Increasing Occupational Health and Safety in Workplaces (pp. 30-47). Cheltenham,
UK: Edward Elgar.
Accident Underreporting 2
Accident Underreporting in the Workplace
In his speech at the XXth World Congress on Safety and Health at Work, ILO Director-General Guy
Ryder stated that “work claims more victims every year than war” (World Congress, 2014; p. 1). One
worker dies from a work-related accident or disease and 160 workers experience a work-related
accident every 15 seconds. Indeed, the statistics for work-related injury and illness are staggering and,
unfortunately, on the rise. In 2017, three years following Ryder’s address, estimates of fatal
occupational incidents and diseases worldwide increased from 2.3 million workers to some 2.78 million
(ILO, 2017). Additionally, there are nearly 374 million non-fatal work-related injuries and illnesses
annually worldwide, with many of these prompting extended work absences. With the increase in such
workplace incidents also comes greater financial burden: the cost of illnesses, injuries, and deaths
combined reached 3.94 percent of global GDP, or $2.99 trillion (ILO, 2017). In the United States alone,
approximately 2.9 million nonfatal workplace injuries and illnesses were reported (Bureau of Labor
Statistics, 2017), representing a price tag of over $1 billion per week for U.S. employers (Liberty Mutual
Workplace Safety Index, 2018).
Despite these sobering statistics, research also increasingly indicates that these statistics may
greatly underrepresent the frequency of non-fatal occupational injuries (e.g, Hämäläinen, Takala, &
Saarela, 2006; Leigh, Marcin, & Miller, 2004; Lowery, et al., 1998; Petitta, Probst, & Barbaranelli, 2017;
Probst, Brubaker, & Barsotti, 2008; Probst, 2015; Probst & Estrada, 2010; Probst & Graso, 2013;
Rosenman, et al., 2006). While there are likely numerous contributing factors to this underestimation,
perhaps the most prominent include individual-level under-reporting (i.e., employees failing to report
work-related illnesses and injuries to their employer) and organizational-level under-reporting (i.e.,
organizations neglecting to accurately report employee illnesses and injuries to regulatory surveillance
authorities). Figure 1 illustrates how these contributing factors can distort the statistics provided by
national and international surveillance systems, thus reflecting only the tip of the proverbial iceberg.
Accident Underreporting 3
Such injury surveillance methods can be implicated at the employee- and/or organizational-level (Petitta
et al., 2017; Probst & Graso, 2011; Weddle, 1996).
The purpose of this chapter is to examine the organizational and psychosocial factors linked to
the under-reporting of workplace accidents. For this purpose, a workplace accident is defined as any
unplanned and uncontrolled event that led to injury to persons, damage to property/plant/equipment,
or some other loss to the company. Thus, we expand beyond a sole focus on injury under-reporting to
also include other significant events that organizations would typically expect their employees to report
(i.e., damage to property or equipment). Below, we first provide a selected review of regularly utilized
processes that organizations and countries have implemented for reporting, investigating, and tracking
employee injuries and illnesses. We then define the concept of accident under-reporting and review
research on the frequency of and organizational and individual correlates of under-reporting. Lastly, we
propose several potential interventions organizations can use to improve the accuracy of accident
reporting as well as propose directions for future research in this area.
Investigating and Reporting Accidents in the Workplace
From a seemingly innocent near-miss incident to the death of a worker, accidents range greatly
in severity, but are generally required to be reported and investigated if they meet certain criteria. In
the hopes in preventing (or minimizing) future incidents, investigations aim to determine the individual,
organizational and/or job-related factors relevant in the accident’s cause. The following section begins
by describing the standard reporting process organizations use and subsequently examines the
procedure of reporting work-related accidents to the proper governmental regulatory authority.
Investigating and Reporting Accidents to Employers
Although accident investigations are conducted in response to an incident, organizations must
be proactive in establishing guidelines and procedures well before they are required. Moreover,
organizations should have an established protocol where an accident investigation team can properly
Accident Underreporting 4
examine the accident from varying perspectives. Such a team is typically comprised of any witnesses,
the injured worker’s immediate supervisor, the company designated safety officer, and an employee
representative elected by the workers to represent them (WA Labor and Industries, 2015). Thus, the
accident investigation team should include an intentionally diverse team that amplifies the objectivity
and diligence of the investigation procedure.
The initial step following an injury is attending to the affected employee(s). Following medical
attention, the preliminary phase of an investigation can commence. Analogous to a fact-finding mission
(Geller, 2016), this stage should include a thorough examination where the accident scene is secured,
photographed, and witnesses interviewed to answer questions such as the following (WA Labor and
Industries, 2015):
Where and when did the accident occur?
Who was present?
What was the employee doing prior to and at the time of the accident?
What occurred during the accident?
After the accident scene is secured and data collected, the accident investigation necessitates
establishing the chain of events leading to the accident and identification of the root causes of the
accident. Causal factors could involve individual (e.g., lack of safety skills, low safety motivation, and/or
distorted risk perceptions), job (e.g., ergonomic factors, workload, and other environmental
characteristics), and/or organizational variables (e.g., poor communication, few resources, a poor safety
climate, and/or lack of safety leadership; Christian, Bradley, Wallace, & Burke, 2009). It is additionally
critical to acknowledge there are both proximal and distal causes of an accident (Christian, et al., 2009).
Examples leading to this failure may include a lack of employee safety knowledge and/or safety
motivation (proximal cause) or this lack of knowledge and/or motivation could in turn be traced to, say,
poor organizational safety climate, a lack of training provided, and/or the employees’ general propensity
Accident Underreporting 5
to make risky behaviors (distal causes). Notably, an accident is often not attributable to a single cause,
but rather a confluence of factors (Christian, et al., 2009; Geller, 2016).
Following the identification of the accident’s proximal and distal causes, the third and final stage
of the accident investigation is the development and implementation of recommendations to prevent or
minimize such events in the future. Results of the investigation may reveal changes needed on the part
of the employee (e.g., attendance at safety training), the job task environment (e.g., installation of
automatic shut-off switches), and/or the organization itself (e.g., long-term efforts to improve safety
climate). Although not technically part of an accident investigation, a formal attempt should be made to
collect follow-up information concerning the effectiveness of the implemented recommendations at
improving the individual, job, and/or organizational risk factors identified during the investigation.
As with accidents, a similar process should be used in investigating and reporting near-miss
incidents. The rationale behind near miss reporting systems is centered around the injury triangle
model. Initially proposed by Heinrich (1931), the model has two assumptions. The first of which is that
for every one serious accident or death, there are roughly 30 accidents with lost days, and 300 near
misses. Some debate surrounds the precise ratio of near-miss to incurred accident. For instance, Bird
and Germain (1996) estimated for every 1 death or catastrophic property loss, there are 30 minor
injuries and/or property damage incidents, and 600 near-misses. Regardless of the exact ratios, the
underlying principles and assumptions are alike (Bellamy, 2015; Nielsen, Carstensen, & Rasmussen,
2006). Together, these models can be understood as “iceberg” models; that is, severe safety incidents
are often the most evident, but numerous minor and potential incidents lurk beneath the surface.
The second assumption is that near-misses and severe injuries share equivalent underlying
causal processes. Near-misses and accidents are argued to merely differ by the slightly varying
circumstances surrounding each event. It is therefore critical to attend as diligently to the
reporting/investigation of near misses as it is actual accidents. Moreover, considering the significantly
Accident Underreporting 6
larger number of near-misses relative to the number of accidents, near-misses provide many low cost,
injury-free opportunities to learn from the conditions in which they occurred. Necessary precautions
implemented before these circumstances can prevent a full-blown lost-time injury or worse (Barach &
Small, 2000; Bellamy, 2015; Nielsen et al., 2006). Despite the practicality of near-miss
reporting/investigation, individual and organizational responses to near-misses unfortunately typically
lack the same sense of urgency as an actual injury. Thus, employees often bypass near-miss reporting
processes and organizations seldom conduct thorough investigations of such incidents.
Organizational Reporting of Accidents to Regulatory Authorities
Accurate surveillance estimates require not only that employees report incidents to their
companies, but also that organizations accurately report workplace injuries and illnesses to the
appropriate governmental regulatory authorities. Such reporting plays an important role, as they are
used for purposes such as performance measurement, inspection targeting, resource allocation, safety
standards development, the identification of high-risk and low-hazard industry sectors, and compilation
by national surveillance programs (OSHA, 2018).
In the United States, the Occupational Safety and Health Administration (OSHA) maintains the
reporting and recordkeeping system to monitor job-related injuries/illnesses. Established by the
Occupational Safety and Health Act of 1970, OSHA facilitates a nationwide surveillance system through
annual logs of workplace injuries/illnesses required from businesses with 10 or more employees.
Records of the log data serve three primary functions: summaries are posted for employees’ viewing,
they are maintained for 5 years in case inspection is requested by OSHA and/or any state regulators, and
they are used to compute injury rates by employer size, industry, and other various classifications.
Outside of the U.S., a similar regulatory body is used in the European Union. Known as the
European Agency for Safety and Health at Work, it is the primary bureaucratic body in developing
guidelines for occupational injury surveillance. Although each EU member country has its own
Accident Underreporting 7
distinctive reporting requirements, there is some standardization directed by the Framework Directive
on Health and Safety in the Workplace, where all accidents leading to an absence of more than four
calendar days must be integrated in the European Statistics on Accidents at Work (ESAW) database
(Eurostat, 2016). Unfortunately, many developing countries largely lack any regulation in enforced
occupational injury monitoring (Hämäläinen et al., 2006). The dearth of monetary resources is the most
commonly cited reason to blame, and with the absence and/or unreliability of data, it is difficult to
obtain accurate estimates of occupational accident rates globally (Hämäläinen et al., 2006).
Definition and Prevalence of Accident Under-reporting
Even when extensive reporting requirements exist at both the organizational and national
levels, research increasingly suggests these numbers represent significant undercounts of the true
extent of work-related injuries and illnesses. Below we define the concept of under-reporting and
examine its prevalence and potential causes.
Organizational-Level Accident Reporting
Definitional terms. While specific reporting requirements differ from country to country, in the
U.S. a reportable event (i.e., one that must be recorded in the Log of Workplace Illnesses and Injuries
and reported to OSHA) is any work-related injury or illness that results in: death, loss of consciousness,
days away from work, restricted job duty or transfer, or medical treatment beyond first aid (Bureau of
Labor Statistics, 2005). Such logged events are referred to as a recordable events. A standardized
incidence rate known as the organization’s recordable rate can then be calculated using the following
equation: (N/EH) x 200,000, where N = the number of recordable events, EH = total hours worked by all
employees during the calendar year and 200,000 = base number of hours for 100 equivalent full-time
workers (i.e., 40 hr./wk, 50 wk./yr.). The resulting standardized incidence rate (i.e., the number of
annual injuries per 100 workers) can then be used to compare injury rates across companies and
industries regardless of number of employees or hours worked (Bureau of Labor Statistics, 2008).
Accident Underreporting 8
Unfortunately, research indicates that not all events meeting the definition of a reportable
event are actually logged. Such events are called unreported events. Organizational under-reporting can
then be said to occur when there is a discrepancy between the number of reportable events and the
number of recordable events appearing in the OSHA Log. Thus, organizational under-reporting is defined
as a function of both (1) the total number of experienced reportable events and (2) the number actually
reported to the regulatory authority (i.e., recordables). As the discrepancy between reportable and
recordable events increases, organizational-level under-reporting can be said to increase.
Prevalence of organizational under-reporting. Precise estimates of organizational under-
reporting vary; however, several studies have documented its prevalence (see Glazner et al., 1998; Leigh
et al., 2004; Pransky, Snyder, Dembe, & Himmelstein, 1999). For example, in a study conducted during
the construction of Denver International Airport, actual injury rates were found to be more than double
the published national rates for that industry (Glazner et al., 1998). Similar rates of underreporting have
been suggested by comparing figures between work-related injuries/illnesses (e.g., Worker’s
Compensation claims or medical records) and national data from the Bureau of Labor Statistics (BLS)
Survey of Occupational Injuries and Illnesses (Leigh et al., 2004; Pransky et al., 1999).
A more comprehensive study by Rosenman et al. (2006) compared figures from all individuals
and companies in Michigan sampled during three years of the annual BLS survey against the
injury/illness data reported to other Michigan databases (e.g., workers’ compensation) by the same
individuals and companies. This comparison revealed that the national surveillance system failed to
accurately reflect up to 68 percent of work-related injuries and illnesses. In a later study, Probst,
Brubaker, and Barsotti (2008), using a sample of 38 construction contractors, revealed nearly 78 percent
of all incidents meeting the definition of a recordable event went unreported in the contractors’ OSHA
Accident Underreporting 9
Given these findings, even the most conservative estimates appear to indicate that half of all
reportable events are not accurately communicated to the appropriate regulatory authority.
Individual-level Under-reporting
Definition and operationalization. Individual-level under-reporting is defined similarly as
organizational under-reporting in that it involves a comparison of the number of experienced workplace
injuries and illnesses that are reported to the company to the number of actual experienced injuries and
illnesses. Thus, as with organizational under-reporting, as the discrepancy between the number of
reported and experienced accidents increases, individual-level under-reporting increases.
Despite their similar definitions, precise measurement of under-reporting at the individual-level
is more challenging than at the organizational-level. At the organizational-level, the number of events
reported to the regulatory authorities (i.e., recordable events) can be found in the company’s OSHA
logs. These data can then be compared to workerscompensation data and/or to medical records data
providing the number of reportable events. Thus, relatively objective figures of reportable vs. recordable
events can be found.
On the other hand, it is much more difficult to accurately capture the number of events that
employees should have reported to their organization. First, the criteria for what constitutes a
reportable event varies from company to company. In some companies, there is a “report everything”
policy which includes close calls, near misses, and unsafe conditions and/or behaviors. Others require
only that actual injuries be reported. Thus, there is no standardized measure of what constitutes a
reportable event as there is at the national level.
More problematic, even if one defines a reportable event using the same criteria as the OSHA
recordable events, there is no objective measure of what actually gets reported to the employer (i.e.,
reported events). As noted earlier, at least half of all reportable events do not make it into the official
Accident Underreporting 10
log of workplace illnesses and injuries. Thus, relying on these logs would severely underestimate the
number of events that employees had reported to their organization.
As a result, researchers typically rely upon self-report data from employees to estimate the
number of reported and experienced events. For example, Probst and Estrada (2010) used a free-recall
measure derived from Smecko and Hayes (1999), which asked employees to indicate how many safety
incidents they experienced and reported and how many they experienced but did not report to their
supervisor over the past 12 months. A safety incident could include any unplanned event that led to
injury, property damage, and/or other loss to the company. Together, these two items allow for a
comparison of the total reportable events (reported and unreported) to the number eventually
reported. A newer recognition-based measure (Probst and Graso, 2013) calculates experienced and
reported events based on the U.S. Bureau of Labor Statistics’ Occupational Injury and Illness
Classification System (OIICS; BLS, 2012) by asking employees to indicate whether they experienced and
subsequently reported exposure to 17 unique events (e.g., slip, trip, fall, hit by object, electrical shock,
etc.). By comparing the levels of experienced and reported events, the extent of underreporting can be
Unfortunately, reliance on self-report safety data can also raise methodological concerns. First,
self-report data can be inaccurate simply due to an inability to correctly recall safety incidents. For
example, the literature suggests that many minor accidents might be forgotten due to extended recall
periods. Additionally, self-report measures may also be misleading due to impression management
goals of the employee.
Prevalence of individual-level underreporting. Given the measurement challenges noted above,
precise estimates of individual-level underreporting are difficult. Nevertheless, research (e.g., Probst and
Estrada, 2010; Probst, Barbaranelli, & Petitta, 2013; Probst & Graso, 2013) indicates it may be quite
pervasive with estimates from those studies suggesting 57–80% of all accidents experienced by
Accident Underreporting 11
employees go unreported to their company. More recently, data from a large public transit agency
(Byrd, Gailey, Probst, & Jiang, 2018) revealed 77% of experienced accidents were not properly reported.
Individual and Organizational Correlates of Reporting Behavior
Despite the challenges in determining the exact pervasiveness of under-reporting, there is a
general consensus that both organizational- and individual-level under-reporting exists. Thus, it is
important to understand the causes and consequences of such under-reporting.
At the organizational-level, clearly organizations with high levels of accident under-reporting do
not have fewer accidents; they just have fewer reported accidents. While superficially having fewer
apparent accidents may have short-term benefits (e.g., a lower workers’ compensation loss rate),
organizations incur substantial costs associated with the long-term health and safety of their employees,
as well as costs related to this failure to address the source of work-related injuries or accidents (Pransky
et al., 1999). These costs may be compounded with government-imposed penalties and fines stemming
from such organizational under-reporting.
Similarly, the costs associated with accident under-reporting extend to the employee. Failure to
utilize worker’s compensation presents an enduring consequence of under-reporting and places the
financial burden of medical claims solely on the employee. Subsequently, this financial responsibility
may result in injuries going untreated, resulting in greater problems for the employee and potentially
their coworkers, e.g., decline in productivity due to attending to work ill or injured (Gallagher & Myers,
1996; Loeppke et al., 2003).
Considering the prevalence of under-reporting and the serious nature of its consequences, it is
vital to identify factors that predict and contribute to individual- and organizational-level under-
reporting. Table 1 summarizes a variety of psychosocial factors that have been theorized and/or
empirically-demonstrated to be predictive of under-reporting.
Predictors of individual-level under-reporting
Accident Underreporting 12
The majority of under-reporting research has focused on root causes for employee under-
reporting rather than organizational-level under-reporting. Furthermore, many of the studies that
assess correlates of individual under-reporting do not actually measure under-reporting as discussed in
this chapter. Rather, many researchers simply gather descriptive data reflecting employee justifications
for not reporting an accident to their employer. However, this approach results in little empirical
analysis of the extent to which these variables are actually predictive of discrepancies between reported
and experienced events.
Despite the aforementioned barriers, researchers have documented numerous predictors of
individual-level under-reporting: demographic characteristics such as age, organizational tenure, and
employment status (Palali & van Ours, 2017; Probst, Petitta, Barbaranelli, & Lavaysse, 2018);
personality characteristics, particularly, consideration of future safety consequences (Probst, Graso,
Estrada, & Greer, 2013); level of trust in the employee-employer relationship (Carmeli & Gittell, 2009);
perceptions of an organization’s workplace and gender climate (Tei-Tominaga & Nakanishi, 2018; Austin,
Probst, Petitta, & Barbaranelli, 2018); safety-related moral disengagement (Petitta et al., 2017);
perceived lack of management responsiveness (Clarke, 1998); fear of reprisals or loss of workplace perks
and pay incentives (Webb, Redman, Wilkinson, & Sanson-Fisher, 1989; Pransky et al., 1999; Sinclair &
Tetrick, 2004); and an acceptance that injuries are commonplace in certain lines of work (Pransky et al.,
1999). Additionally, employees may have issues classifying whether or not an incident meets the
definition of an accident, which may stem from a lack of adequate safety training and/or improper
attributions regarding the event’s cause (Pransky et al., 1999). Furthermore, laborious (Glendon, 1991)
or punitive reporting systems can dissuade employees from reporting an incident. Indeed, if employees
anticipate that they will not receive just treatment from their employer, the likelihood of them reporting
an incident decreases substantially (Weiner, Hobgood, & Lewis, 2008).
Accident Underreporting 13
Misguided safety incentive systems also may help explain the prevalence of individual-level
under-reporting. Although well-intended, many serve to reward employees for being accident-free
instead of safety compliant. These systems inadvertently encourage employees to under-report
experienced accidents in order to preserve their ability to receive company-administered safety rewards
(Pransky et al., 1999). These apparent social pressures, combined with work group norms, serve as
deterrents for employee to report experienced accidents to their supervisors (Sinclair & Tetrick, 2004).
In a similar fashion, research (Jiang, Probst, Benson, & Byrd, 2018) also indicates that employees who
are victims of physical and psychological aggression at work tend to not only experience more injuries at
work, but also underreport those experiences, potentially out of fear of retaliation.
Individual perceptions of an organization’s safety climate (also referred to as psychological
safety climate; Clarke, 2009) and the degree to which supervisors enforce safety policies represent
additional predictors of individual under-reporting. Specifically, under-reporting diminished when
employees perceived a positive organizational safety climate (Probst & Estrada, 2010). Furthermore,
when supervisors continually enforced safety policies, employees both experienced fewer accidents and
more fully reported those that were experienced. In contrast, a poor safety climate and/or lax
enforcement was associated with a higher ratio of unreported to reported accidents (greater than 3:1).
Production pressure represents an additional factor that may have a detrimental impact on
employee health and safety (e.g., Landsbergis, Cahill, and Schnall, 1999; McLain & Jarrell, 2007). In
support of this, Probst and Graso (2013) examined the extent to which production pressure was related
to employee attitudes and behaviors regarding accident reporting. Results suggest that employees who
perceived higher levels of production pressure reported increased negative attitudes towards the
reporting of accidents. Additionally, as perceptions of production pressure increased, rates of accident
under-reporting concurrently increased.
Accident Underreporting 14
An apparent motivator for accident reporting may stem from consideration of potential
negative consequences of reporting such events. Indeed, Probst and Estrada (2010) found that 64% of
all respondents indicated they had experienced at least one negative consequence as a result of
reporting an accident. These consequences ranged from poor interpersonal treatment (e.g., being
blamed for the incident) to adverse job performance outcomes (e.g. poor performance review). Such
concerns also increase fears regarding the sustainability of valued job incentives or one’s position within
an organization (i.e. job security; Mitch & Laumann, 2016; Palali & van Ours, 2017; Probst, Barbaranelli,
and Petitta, 2013). For example, Probst et al. (2013) found that job insecurity (i.e., fear of job loss) was
not only a significant predictor of the prevalence of accidents/injuries, but also of accident under-
reporting decisions. Similarly, Probst and Graso (2013) found that prior experience with negative
consequences of reporting accidents resulted in increased reluctance to fully report experienced
Predictors of organizational-level under-reporting
Though research mainly focuses on individual-level under-reporting and often include variables
operating at the organizational-level (e.g. production pressure; safety climate), there are unique factors
that are strictly associated with organizational-level under-reporting. For example, complex or
bureaucratic safety management systems, pervasive fear of litigation and negative publicity, as well as
economic pressures, government penalties for safety violations, higher insurance premiums and
collective bargaining agreements, represent disincentives for accurate organizational reporting of
experienced accidents and injuries (Barach & Small, 2000; Conway & Svenson, 1998; Jeffcott, Pidgeon,
Weyman, & Walls, 2006; Milch & Laumann, 2016; Palali & van Ours, 2017; Zacharatos & Barling, 2004).
Influenced by such disincentives, Zahlis and Hansen (2005) argue that there is a continual
disconnect between organizational measures (OSHA recordables) and valued outcomes (workers’
compensation costs and expenses). Furthermore, these discrepancies highlight an over-reliance on poor
Accident Underreporting 15
(often lagging) indicators of organizational safety and the high stakes attached to an organization’s
OSHA injury rate.
Probst, Brubaker, and Barsotti (2008) utilized medical records data provided by an Owner-
Controlled Insurance Program, finding significant differences between the number of reported injuries
in OSHA logs and the number actually experienced. Specifically, for every 3 injuries that were properly
recorded, an additional 11 went unrecorded in the log. Furthermore, this rate of under-reporting was
significantly predicted by the organizational safety climate, such that organizations with a poor safety
climate had a significantly higher rate of unreported injuries than organizations with a positive safety
climate (17 versus 3 per 100 workers). Interestingly, there was no significant difference in the injury
rates actually reported to OSHA (3.98 vs. 3.64), indicating that organizations with poor and positive
safety climates reported comparable injury rates. Yet, those reports were far more distorted (i.e.,
inaccurate) for the companies with poor safety climates. These results further highlight the unreliability
of using OSHA logs as the primary metric of safety behaviors within an organization.
Addressing their findings, Probst and colleagues (2008) acknowledged that the discrepancies
could possibly be due to intentional manipulation of the numbers. However, it is equally plausible that
such discrepancies could be attributable to the fact organizations with a poor safety climates may not
devote adequate resources towards accurately tracking recordables; may not provide proper training for
safety personnel regarding proper identification of recordable injuries; and may have implicit incentive
systems that reward managers and administrators for coding injuries as something other than a
recordable. These and other findings (e.g. Probst, 2015; Zadow, Dollard, Mclinton, Lawrence, & Tuckey,
2017), emphasize that safety climate represents a significant predictor of organizational-level under-
Implications and Future Research Needs
Accident Underreporting 16
A growing body of research has documented and identified numerous individual and
organizational factors associated with accident under-reporting. Although the vast majority of this
research has considered these factors in isolation from each other, collectively the findings from the
extant literature indicate the need to develop comprehensive multi-level models of accident reporting
behavior including variables operating at the individual-, workgroup-, organizational-, and national-
levels that contribute to the problem of under-reporting.
Workplace interventions (and evaluations of the extent to which these are successful) are also
needed to improve the accuracy of accident reporting. While recommendations for improving safety
incentive systems and reporting systems have been suggested, the effectiveness of such interventions
are rarely empirically tested. For example, Reason (1997) argued that there are 5 preconditions
necessary for organizations to report accurately: (1) indemnity against disciplinary action, (2)
confidentiality of reporting, (3) separating the agency who collects and analyzes the data from the
regulatory authority, (4) rapid and useful availability of feedback, and (5) ease in using the reporting
system. Based on the literature review summarized in Table 1, many of these same recommendations
would apply at the individual-level: (1) no-fault reporting where accident investigations are seen as fact-
finding as opposed to fault-finding; (2) confidential and/or anonymous reporting; (3) visible and positive
organizational response to reports of hazardous situations and/or injuries; and (4) a simple
straightforward reporting system.
It is important to also understand what types of injuries and illnesses go unreported, so as to
better target interventions. The inclusion of actual loss data in future studies (i.e., workers’
compensation claims rates and costs) would allow for a more in-depth analysis of the kinds of injuries
that are being unreported in workplace injury and illness logs. Similarly, in self-report surveys,
researchers should determine if employees fail to report certain types of accidents more often than
others. For example, in their study of copper miners, Probst and Graso (2013) used open-ended
Accident Underreporting 17
questions to ask employees how many total accidents they had experienced and reported and how
many they had experienced but not reported. They then asked workers to indicate how many of the
accidents within each category had resulted in lost-time, first-aid, no injury, and property/equipment
damage. Using this metric, they found that nearly 52% of all accidents went unreported. However, the
highest rates of under-reporting occurred with events requiring only first aid (89% unreported) or that
resulted in a self-assessment of “no injury” (61% unreported). On the other hand, only 1/3 of all lost-
time injuries and 24% of all damage to property or equipment went unreported. This would seem to
indicate that employees are more willing to report more severe injuries (particularly those requiring
time off from work) and those resulting in property damage, perhaps due to the higher visibility of such
events. On the other hand, smaller, less noticeable events that apparently causelittle or no” harm tend
not to get reported. This comports with a study conducted by Nielsen, Carstensen, and Rasmussen
(2006) which asked metal workers to indicate their willingness to report several types of hypothetical
events. In that study, employees indicated they would be most willing to report lost-time incidents
followed by minor injuries and then near misses.
Frequently cited national and international injury, illness, and fatality statistics can appear
staggering; yet, a large and growing body of research suggests these data may only represent the tip of
the iceberg. This chapter presented evidence of major discrepancies between accidents experienced by
employees, what gets reported to organizations, and ultimately what organizations report to the
regulatory authorities. Using conservative estimates, it appears that for every 1 injury captured in the
national surveillance system data, at least 2 incidents were reported to organizations, and at least 4
were actually experienced by employees. Extrapolating from the 2.9 million officially reported work-
related illnesses and injuries in the United States in 2017, this would imply that there were an additional
13 million events hidden below the tip of that iceberg. Given the prevalence of such under-reporting, we
Accident Underreporting 18
reviewed research on empirically-established individual and organizational correlates of under-reporting
and discussed the need for more comprehensive multilevel research examining the interplay among
these correlates, as well as more research on practical interventions that organizations can utilize to
improve the accuracy of accident reporting.
Accident Underreporting 19
Austin, C. T., Probst, T. M., Petitta, L., & Barbaranelli, C. (2018). Gender climate and accident
underreporting: The role of gender congruence. Manuscript in preparation.
Barach, P., & Small, S. D. (2000). Reporting and preventing medical mishaps: Lessons from non-medical
near miss reporting systems. British Medical Journal, 320, 759-763.
Bellamy, L. J. (2015). Exploring the relationship between major hazard, fatal and non-fatal accidents
through outcomes and causes. Safety Science, 71, 93-103.
Bird, Jr., F. E., & Germain, G. L. (1996). Practical loss control leadership. Loganville, GA: International Loss
Control Institute.
Bureau of Labor Statistics. (2005). Occupational safety and health definitions. Retrieved from
Bureau of Labor Statistics, 2008. Workplace Injury and Illness Summary. United States Department of
Labor, Washington, DC.
Bureau of Labor Statistics, 2012. Occupational Injury and Illness Classification Man- ual, Available at:
Bureau of Labor Statistics (2017). Workplace injuries and illnesses in 2016. Retrieved from
Byrd, J., Gailey, N. J., Probst, T. M., & Jiang, L. (2018). Explaining the job insecurity-safety link in the
public transportation industry: The mediating role of safety-production conflict. Safety Science,
106, 255-262.
Carmeli, A., & Gittell, J.H. (2009). High-quality relationships, psychological safety, and learning from
failures in work organizations. Journal of Organizational Behavior, 30, 709729.
Accident Underreporting 20
Christian, M. S., Bradley, J. C., Wallace, J. C., & Burke, M. J. (2009). Workplace safety: A meta-analysis of
the roles of person and situation factors. Journal of Applied Psychology, 95, 1103-1127.
Clarke, S. (1998). Organizational factors affecting the incident reporting of train drivers. Work & Stress,
12, 6-16.
Clarke, S. (2009). Accidents and safety in the workplace. In: Cartwright, S., Cooper, C.L. (Eds.), The Oxford
handbook of organizational well-being (pp. 31-54). Oxford University Press, Oxford, UK.
Conway, H., & Svenson, J. (1998). Occupational injury and illness rates, 1992-1996: Why they fell.
Monthly Labor Review, 34-58.
Eurostat (2016). Accidents at Work Statistics. Luxembourg: Office for Official Publications of the
European Communities.
Gallagher, R. M., & Myers, P. (1996). Referral delay in back pain patients on worker’s compensation:
Cost and policy implications. Psychosomatics: Journal of Consultation Liaison Psychiatry, 37, 270-
Geller, E. S. (2016). The Psychology of Safety Handbook. CRC press.
Glazner J. E., Borgerding J., Lowery J. T., Bondy J, Mueller K. L., & Kreiss K. (1998). Construction injury
rates may exceed national estimates: Evidence from the construction of Denver International
Airport. American Journal of Industrial Medicine, 34, 105-112.
Glendon, A.I. (1991). Accident data analysis. Journal of Health and Safety, 7, 5-24.
Hämäläinen, P., Takala, J., Saarela, K. L. (2006). Global estimates of occupational accidents. Safety
Science, 44, 137-156.
Heinrich, H. W. (1931). Industrial accident prevention. New York, NY: McGraw-Hill.
International Labor Organization (2017) Safety and Health at Work. Retrieved from
Accident Underreporting 21
Jeffcott, S., Pidgeon, N., Weyman, A., & Walls, J. (2006). Risk, trust and safety culture in U.K. train
operating companies. Risk Analysis, 26, 11051121.
Jiang, L., Probst, T. M., Benson, W., & Byrd, J. (2018). Voices carry: Effects of verbal and physical
aggression on injuries and accident reporting. Accident Analysis and Prevention, 108, 190-199.
Landsbergis, P. A., Cahill, J., & Schnall, P. (1999). The impact of lean production and related new systems
of work organization on worker health. Journal of Occupational Health Psychology, 4, 108-130.
Leigh, J.P., Marcin, J.P., & Miller, T.R. (2004). An estimate of the U.S. government’s undercount of
nonfatal occupational injuries. Journal of Occupational and Environmental Medicine, 46, 10-18.
Liberty Mutual Insurance Safety Index. (2018, May 08). Workplace Injuries Cost U.S. Companies Over $1
Billion Per Week. Retrieved from
Loeppke, R., Hymel., P. A., Loafland, J. H., Pizzi, L. T., Konicki, D. L., Anstadt, G. W., Baase, C., Fortuna, J.,
& Scharf, T. (2003). Health-related workplace productivity measurement: General and migrane-
specific recommendations from the ACOEM expert panel. Journal of Occupational and
Environmental Medicine, 45, 349-359.
Lowery, J.T., Borgerding, J.A., Zehn, B., Glazner, J.E., Bondy, J., & Kreiss, K. (1998). Risk factors for injury
among construction worker at Denver International Airport. American Journal of Industrial
Medicine, 34, 105-12.
McLain, D. L., & Jarrell, K. A. (2007). The perceived compatibility of safety and production expectations
in hazardous occupations. Journal of Safety Research, 38, 299-309.
Milch, V., & Laumann, K. (2016). Interorganizational complexity and organizational accident risk: A
literature review. Safety Science, 82, 9-17.
Accident Underreporting 22
Nielsen, K. J., Carstensen, O., & Rasmussen, K. (2006). The prevention of occupational injuries in two
industrial plants using an incident reporting scheme. Journal of Safety Research, 37, 479 486.
OSHA (2018). Employer rights and responsibilities following a federal OSHA inspection. Retrieved from
Palali, A., & van Ours, J. C. (2017). Workplace accidents and workplace safety: On under-reporting and
temporary jobs. Labour, 31(1), 1-14.
Petitta, L., Probst, T. M., & Barbaranelli, C. (2017). Safety culture, moral disengagement, and accident
underreporting. Journal of Business Ethics, 141(3), 489-504.
Pransky, G., Snyder, T., Dembe, A., & Himmelstein, J. (1999). Under-reporting of work-related disorders
in the workplace: A case study and review of the literature. Ergonomics, 42, 171-182.
Probst, T. M. (2015). Organizational safety climate and supervisor safety enforcement: Multilevel
explorations of the causes of accident underreporting. Journal of Applied Psychology, 100(6),
Probst, T. M., Barbaranelli, C., & Petitta, L. (2013). The relationship between job insecurity and accident
under-reporting: A test in two countries. Work & Stress, 27(4), 383-402.
Probst, T. M. Brubaker, T. L., & Barsotti, A. (2008). Organizational under-reporting of injury rates: An
examination of the moderating effect of organizational safety climate. Journal of Applied
Psychology, 93(5), 1147-1154.
Probst, T. M., & Estrada, A. X. (2010). Accident under-reporting among employees: Testing the
moderating influence of psychological safety climate and supervisor enforcement of safety
practices. Accident Analysis & Prevention, 42(5), 1438-1444.
Probst, T. M., & Graso, M. (2011). Reporting and investigating accidents: Recognizing the tip of the
iceberg. In S. Clarke, C. Cooper & R. Burke (Eds.), Occupational health and safety: Psychological
and behavioral challenges (pp. 7194). Surrey: Gower.
Accident Underreporting 23
Probst, T. M., & Graso, M. (2013). Pressure to produce = pressure to reduce accident
reporting? Accident Analysis & Prevention, 59, 580-587.
Probst, T. M., Graso, M., Estrada, A. X., & Greer, S. (2013). Consideration of future safety consequences:
a new predictor of employee safety. Accident Analysis & Prevention, 55, 124-134.
Probst, T. M., Petitta, L., Barbaranelli, C., & Lavaysse, L. M. (2018). Moderating effects of contingent
work on the relationship between job insecurity and employee safety. Safety Science, 106, 285-
Reason, J. T. (1997). Managing the risks of organizational accidents. Aldershot: UK: Ashgate.
Rosenman, K., D., Kalush, A., Reilly, M. J., Gardiner, J. C., Reeves. M., & Luo, Z. (2006). How much work-
related injury and illness is missed by the current national surveillance system. Journal of
Occupational and Environmental Medicine, 48, 357-365.
Sinclair, R. R., & Tetrick, L. E. (2004). Pay and benefits: The role of compensation systems in workplace
safety.. In J. Barling, & M. Frone (Eds.) Psychology of workplace safety, (pp. 181-201).
Washington, DC: American Psychological Association.
Smecko, T., & Hayes, B. (1999). Measuring compliance with safety behaviors at work. Paper presented
at the 14th annual conference of the Society for Industrial and Organizational Psychology.
Atlanta, Georgia: April.
Tei-Tominaga, M., & Nakanishi, M. (2018). The influence of supportive and ethical work environments
on work-related accidents, injuries, and serious psychological distress among hospital nurses.
International Journal of Environmental Research and Public Health, 15(2), 240.
Washington State Department of Labor and Industries (2015). Washington Department of Labor and
Industries claims management performance audit. Retrieved from
Accident Underreporting 24
Webb, G. R., Redman, S., Wilkinson, C. & Sanson-Fisher, R. W. (1989). Filtering effects in reporting work
injuries. Accident Analysis and Prevention, 21, 115-123.
Weddle, M. G. (1996). Reporting occupational injuries: The first step. Journal of Safety Research, 27,
Weiner, B. J., hobgood, C., & Lewis, M. A. (2008). The meaning of justice in safety incident reporting.
Social Science and Medicine, 66, 403-413.
World Congress on Safety and Health at Work ILO Director-General: 'Work claims more victims than war'
(2014, August 26). Retrieved from
Zacharatos, A., & Barling, J. (2004). High performance work systems and occupational safety. In J.
Barling, & M. Frone (Eds.) Psychology of workplace safety, (pp. 203-222). Washington, DC:
American Psychological Association.
Zadow, A. J., Dollard, M. F., Mclinton, S. S., Lawrence, P., & Tuckey, M. R. (2017). Psychosocial safety
climate, emotional exhaustion, and work injuries in healthcare workplaces. Stress and Health,
33(5), 558-569.
Zahlis, D. F., & Hansen, L. L. (2005). Beware the disconnect. Professional Safety, 50, 18-24.
Accident Underreporting 25
Table 1. Antecedents of individual and organizational under-reporting
Predictors of individual-level under-reporting Predictors of organizational-level under-reporting
Individual variables
Psychological safety climate
Job insecurity
Improper diagnosis or causal attributions
Fear of reprisals
Consideration of future safety consequences
Safety-related moral disengagement
Perceived safety-production conflict
Experienced psychological and physical
Employment status (contingent vs. permanent)
Organizational and work-group variables
Organizational safety climate
Supervisor safety enforcement
Lack of training
Misguided safety incentive programs
Work group norms / peer pressure
Production pressure
Onerous and/or lack of reporting systems
Organizational justice perceptions
Lack of trust in management
Punitive vs. non-punitive reporting
Organizational variables
Organizational safety climate
Collective bargaining agreements
Organizational size
Industry sector
Regulatory and other external variables
Government penalties and fines
Insurance premiums / Experience
Modification Rates
Fear of litigation
Fear of negative publicity exposure
Accident Underreporting 26
Figure Captions
Figure 1. The Accident Reporting Iceberg
Accident Underreporting 27
... Under-reporting can be compromised at two levels. Establishment level under-reporting is occurring when the record of employees' illness and injuries are not recorded in the regulatory body log of work-related illness and injuries while individual level under-reporting is when workers fail to report illness and injuries that happen in the workplace (Eurostat, 2013;Probst, Bettac, & Austin, 2019). A further problem for under-reporting is that the self-employed are hardly report accidents to the regulatory bodies (Weerd et al., 2014). ...
... In construction similar rates of under-reporting likely exist, as workers may perceive injuries as either small or 'part of the job', though many also fear negative consequences in injury reporting (Taylor Moore et al., 2013). Notably, both frontline workers and managers under report personal incidents to significant degrees, suggesting that both individual-level and organisational-level underreporting persists (Probst et al., 2019). The implications of under-reporting are significant, for both industry and individual. ...
Full-text available
Australia’s population is ageing, but with enhanced health prospects and insufficient retirement funds, and industries impacted by a dwindling itinerate manual labour supply, workers will want, and may need, to remain in the workforce for longer. However, as people age, they lose muscular strength, experience a decline in physical and cognitive performance, and are more vulnerable to muscular-skeletal issues caused by repetitive or awkward movement patterns. Consequently, ageing workers in occupations that require sustained physical activities are at increased risk of injury and exacerbated physical decline and may experience ageist discrimination in the workplace that impacts their psychological wellbeing. This research, Enabling an Ageing Workforce, recognises the issues facing the older worker across a range of different workplace contexts and asks the question: How can design and new technologies address the compounding factors of an ageing (working) population and enable older workers to continue to be productive and effective whilst ensuring their personal wellbeing?
... According to the statistics of the International Labor Organization (ILO), approximately 2.3 million workers die each year due to accidents or diseases worldwide. Furthermore, workplace accident costs have been estimated at US$ 2.8 trillion [1]. Every accident or illness stems from the disagreeable events that occur in workplaces [2]. ...
... Estimates of fatal occupational incidents and diseases worldwide amount to over 2.78 million (ILO, 2017). Additionally, almost 374 million non-fatal work-related injuries and illnesses occur annually worldwide, which result in many instances of extended work absences (Probst, Bettac, & Austin, 2019). These global workplace accident statistics suggest that safety can be further improved for the benefit of both workers and employers. ...
Full-text available
Understanding the factors that either facilitate or hinder the performance of specific safety behaviours is impor- tant in developing effective intervention strategies. A questionnaire to identify determinants of safety behaviours for safety–critical workers does not currently exist. This study reports the development and validation of the Safety Behaviour Change Questionnaire (SBCQ) based on the Theoretical Domains Framework (TDF). Following initial questionnaire development, a 3-stage testing procedure was adopted with three independent rail worker samples (totalling 620 participants), with a focus on three separate specific safety behaviours (removing slip/trip hazards, using PPE, safe tool storage). Exploratory factor analysis (EFA) was used for the identification of the underlying structure of the initial set of items. Confirmatory factor analysis (CFA) was undertaken to generate the model of best fit at the calibration and validation stages. The final version of the SBCQ consisted of 13 factors and 26 items. Subsequent analysis of psychometric invariance confirmed the stability of the model factor struc- ture across three distinct research sub-samples. These initial results suggest that the SBCQ demonstrates reliable, stable and valid properties, and that it can be utilised by safety managers and practitioners to guide the design of safety interventions for a range of safety behaviours.
Full-text available
Introduction: Solid waste workers are exposed to an extended variety of occupational hazards. Among these hazards is the infection from hepatitis A, B, or C viruses (HAV, HBV, or HCV). This relationship has been the study subject of many researchers around the world, given that the infection of hepatitis viruses is a significant cause of morbidity and a socio-economic burden. Solid waste handlers are usually at significant risk for multiple injuries and illnesses, including HIV and hepatitis, due to waste exposure to contaminated needles or sharp items that may contribute to the spread of the disease. A research in Brazil revealed that 12.8% of HBV exposure is prevalent in municipal solid waste handlers. Objectives: To assess the prevalence rate of hepatitis C among the solid waste handler in selected areas and associate the findings with selected demographic variables. Material and methods: This study was used as a cross-sectional research design. Hundred solid waste handlers participated in the study. The prevalence of hepatitis C was checked by the blood sampling and use method: HCV Ab Rapid Test. Data were analyzed using the IBM Corp. Released 2016. IBM SPSS Statistics for Windows, Version 24.0. Armonk, NY: IBM Corp. Qualitative variables were described as numbers and percentages. Chi-square exact test was used for comparison between groups, as a quantitative variable was described as mean (± SD) and median. Results: 10 (10%) of the waste handlers were reactive to hepatitis C virus, and 90 (90%) of the waste handlers were non-reactive to hepatitis C virus. The mean was 1.92 ± 0.27 for the prevalence of hepatitis C among solid waste handlers. Conclusion: A high prevalence of hepatitis C is revealed, particularly in people who have more working experience, exposure, and who do not use personal protective equipment while working around hepatitis C infected people. It is recommended that all the solid waste handlers use proper personal protective equipment, go for routine health check-ups, and should be trained on handling waste to reduce morbidity.
Full-text available
The healthcare industry in Japan has experienced many cases of work-related injuries, accidents, and workers’ compensation claims because of mental illness. This study examined the influence of supportive and ethical work environments on work-related accidents, injuries, and serious psychological distress among hospital nurses. Self-reported questionnaires were distributed to nurses (n = 1114) from 11 hospitals. Valid responses (n = 822, 93% women, mean age = 38.49 ± 10.09 years) were used for analyses. The questionnaire included items addressing basic attributes, work and organizational characteristics, social capital and ethical climate at the workplace, psychological distress, and experience of work-related accidents or injuries in the last half year. The final model of a multivariate logistic regression analysis revealed that those who work less than 4 h of overtime per week (OR = 0.313), those who work on days off more than once per month (OR = 0.424), and an exclusive workplace climate (OR = 1.314) were significantly associated with work-related accidents or injuries. Additionally, an exclusive workplace climate (OR = 1.696) elevated the risk of serious psychological distress. To prevent work-related compensation cases, which are caused by these variables, strengthening hospitals’ occupational health and safety is necessary.
Recent years have witnessed a staggeringly high number of workplace aggressive behaviors as well as employee accidents and injuries. Exposure to workplace aggression is associated with a host of negative psychological, emotional, and physiological outcomes, yet research relating workplace aggression to employee safety outcomes is lacking. This study aims to examine the association between exposure to workplace physical and verbal aggression with workplace injuries and underreporting of accidents and near misses. Furthermore, deriving from social exchange theory, we attempt to reveal an underlying mechanism in the association between workplace aggression and underreporting of accidents and near misses. Finally, borrowing from aggression research on intimate relationships, we compare the relative importance of exposure to physical and verbal aggression on workplace injuries and underreporting. Using survey data from 364 public transportation personnel, we found that both verbal and physical aggression significantly predict workplace injuries as well as underreporting. Moreover, mediation analyses found that the relationship between verbal and physical aggression and underreporting was largely explained by an increase in negative reporting attitudes (rather than decreases in safety knowledge or motivation). Compared to exposure to physical aggression, exposure to verbal aggression best predicted employee underreporting of accidents and near misses. However, physical aggression was a better predictor of injuries than verbal aggression. Given these findings, organizational leaders should strive to foster a safe working environment by minimizing interpersonal mistreatment and increasing employee attitudes for reporting accidents.
Preventing work injuries requires a clear understanding of how they occur, how they are recorded, and the accuracy of injury surveillance. Our innovation was to examine how psychosocial safety climate (PSC) influences the development of reported and unreported physical and psychological workplace injuries beyond (physical) safety climate, via the erosion of psychological health (emotional exhaustion). Self-report data (T2, 2013) from 214 hospital employees (18 teams) were linked at the team level to the hospital workplace injury register (T1, 2012; T2, 2013; and T3, 2014). Concordance between survey-reported and registered injury rates was low (36%), indicating that many injuries go unreported. Safety climate was the strongest predictor of T2 registered injury rates (controlling for T1); PSC and emotional exhaustion also played a role. Emotional exhaustion was the strongest predictor of survey-reported total injuries and underreporting. Multilevel analysis showed that low PSC, emanating from senior managers and transmitted through teams, was the origin of psychological health erosion (i.e., low emotional exhaustion), which culminated in greater self-reported work injuries and injury underreporting (both physical and psychological). These results underscore the need to consider, in theory and practice, a dual physical–psychosocial safety explanation of injury events and a psychosocial explanation of injury underreporting.
Statistics on workplace accidents do not always reflect workplace safety because workers under-report for fear of job-loss if they report having had an accident. Based on an analysis of fatal and non-fatal workplace accidents and road accidents in 15 EU-countries over the period 1995–2012, we conclude that there seems to be cyclical fluctuations in reporting of non-fatal workplace accidents. Workers are less likely to report a workplace accident when unemployment is high. Furthermore, analyzing data from Italy and Spain on both workplace accidents and commuting accidents, we conclude that workers on temporary jobs are likely to under-report accidents. © 2017 CEIS, Fondazione Giacomo Brodolini and John Wiley & Sons Ltd
Previous research has established a link between job insecurity and a myriad of safety outcomes; yet, the explanatory mechanism for this link is unexplored. The purpose of the current study was to explore the role of safety-production conflict (SPC) as a mediator between the relationship of job insecurity and six workplace safety outcomes: behavioral safety compliance, poor accident reporting attitudes, workplace injuries, experienced safety events, unreported safety events, and accident underreporting. Our hypotheses were tested using data from a sample of 389 public transit employees in the United States. Using a bootstrap sampling technique, mediation analyses revealed significant direct and indirect effects (mediation through SPC) of job insecurity on aforementioned workplace safety outcomes. Specifically, higher levels of job insecurity were associated with higher levels of SPC, which, in turn, were associated with detrimental workplace safety outcomes. In the context of improving employee safety, these results suggest that efforts to manage employee perceptions regarding safety-production tradeoffs are of particularly importance in light of today’s pervasive job insecurity during times of global financial crises.
The recent global financial crisis has resulted in heightened levels of employee job insecurity, as well as an increased reliance on a contingent workforce. The purpose of the current study was to examine the conjoint effects of these factors on employee safety-related outcomes. Using survey data from a sample of 1228 employees from a variety of different private and public organizations in Italy, we tested theoretically-derived competing vulnerability and immunity hypotheses regarding the interaction between contingent work and job insecurity. Our results generally supported the vulnerability hypothesis, suggesting that contingent work coupled with job insecurity significantly increase employee risk for poor safety-related outcomes. Specifically, under conditions of job insecurity, contingent workers displayed more adverse safety-related outcomes (e.g., worse safety compliance, safety knowledge, and safety participation) compared to permanent workers. However, the accuracy of their accident reporting was increased under conditions of job insecurity, compared to permanent employees. We discuss these findings in light of important concerns about the safety of contingent workers in the wake of the most recent economic and financial crisis.
A decline in occupational injury and illness rates in the early to mid-1990s is attributable to legislative reforms motivated by increases in workers' compensation payments and a growing awareness of workplace hazards by unions, employers, and the insurance industry.