Psychosocial factors and safety in high-risk industries: A systematic literature review

This systematic review evaluated the evidence between psychosocial factors and safety in high-risk industries.

I think it’s more of a reference paper (one you look up or refer to the tables), than one I can summarise, since most of the findings are captured in tables, but I’ll give it a shot.

40 studies met inclusion criteria.

Providing background:

·?????? “Most large-scale industrial catastrophes (like the Deepwater Horizon oil spill, or Fukushima-Daiichi nuclear disaster) result from a combination of faults in technical arrangements and neglected social structures featuring a workplace”

·?????? They note over time “there has been a fundamental shift in focus from a simple human-technology interaction view to a broader and more holistic way of thinking, emphasizing complex non-linear and nondeterministic interactions and relationships”

·?????? “it has been argued that the term “human error” should be replaced with a term that do not indicate any attributional assumptions to the individual, for instance, “action error” (Mathisen et al., 2017)”

·?????? “accidents cannot be attributed to the behavior of an individual component (i.e., a human error), instead we must examine how interactions between components failed; that is, how the system itself failed”

·?????? To better understand individual behaviour in complex high-risk settings, the research aims to investigate psychosocial factors that increase the risk of errors and accidents (which they say is aligned to Safety-I thinking), and also to explore everyday work and things that went well (aligned to Safety-II thinking)

·?????? Psychosocial factor is neutral and includes work demands, organisational support, rewards, interpersonal relationships and more. Psychosocial hazards are the specific factors that have inherent potential for individual or organisational harm. Psychosocial risk is the likelihood of psychosocial hazards resulting in harm

·?????? The authors grounded the findings primarily using the Demands-Resources (JD-R) theoretical model. I’ve largely skipped the discussion of what these are, but they? are extremely well studied

·?????? Job demands are “those physical, psychological, social, or organizational aspects of the job that require sustained physical and/or psychological (cognitive and emotional) effort or skills”

·?????? Job resources are “those physical, psychological, social, or organizational aspects of the job that are either/ or functional in achieving work goals; reduce job demands and the associated physiological and psychological costs; stimulate personal growth, learning, and development”

·?????? Action errors, as used in this paper are defined as “unintended deviations from plans, goals, or adequate feedback processing, as well as incorrect actions resulting from lack of knowledge”

·?????? They also drew on the distinction between safety compliance (carrying out work in per procedures and accepted/expected practices) and safety participation behaviours (helping coworkers, promoting safety in the workplace, demonstrating initiative and more)

Their conceptual model is shown below:

Note. I’ve had to skip a lot of findings. Any table or data in this summary is likely to have been heavily truncated from the full findings of the paper. The paper is open access though, so you can read it in full yourself.?

Results

Overall they note:

·?????? There is evidence of a link between the exposure to workplace psychosocial factors and safety outputs in high-risk industries

·?????? Stronger links in the extant literature are found between psychosocial factors and safety behaviours (compliance and participation) than between psychosocial factors and safety outputs/events

·?????? Job demand factors are “likely to trigger employees’ health-impairing mental/physical conditions that can constitute a precursor of unsafe behavior”

·?????? Links were found between work-induced psychosocial states (typically in the form of stress or exhaustion) and safety outputs

Discussing the findings, it’s noted that ? of studies came from China, Norway, Hong Kong and the UK (with half from Norway and China); hence, the findings may not be representative globally.

Nevertheless, it’s argued that “Dollard et al. (2014) accentuated the need to account for cross-cultural differences when discussing the subject of workplace psychosocial factors as the comparison of results derived from Western and non-Western countries might not always be meaningful”.

Further, ? of the studies were from construction, oil and gas and mining; again, meaning some caution in extending the findings to other industries.

They said overall, there is “some evidence of a link between the exposure to workplace psychosocial factors (i.e., in the form of job demands and resources) and ensuing employees’ safety violating (or supporting) behaviors in high-risk industries”.

That is, psychosocial factors and demands of the workplace influence consequent behaviours. For instance perceived job insecurity affects job satisfaction and subsequent safety behaviours (self-protective behaviours and the like).

As noted earlier, more research has established links between psychosocial factors and subsequent safety behaviours rather than safety outputs. This is expected, since behaviours are more immediate and temporally linked to preceding psychosocial factors, compared to downstream safety incidents or other output measures.

As an example, only half of the studies that explored psychosocial factors and safety outputs found a significant relationship whereas 80% of studies found a significant relationship between psychosocial factors and safety behaviours.

Most studies involving job demands involved a form of job-induced pressure, like production pressure, work pressure or time pressure; which are linked to safety-critical factors.

Some of the studies found no significant relationships between job-induced pressure and safety-specific effects, suggesting that “job-induced pressure can lead to adverse safety-specific effects that are additionally dependent on, for instance, unique personal, situational, or institutional factors”.

Some researchers have argued that job challenges and job hindrances should be viewed as distinct as they are related to different individual basic psychological needs.

Also, while the majority of studies relied on theoretically sound and valid psychosocial constructs in their methods, we should also consider a wide raft of other relevant factors like work-home conflict, work underload, harassment and perceptions of change which haven’t had the same degree of study.

The study also explored psychosocial factors and “exerted psychosocial states”. I’ve largely skipped this section (e.g. not provided any of the tabulated data). Of interest is that, to date, most of the research in the high-risk domain has focused on the links with job resources and stress at an individual level and less about examining the role of leadership support in mitigating emotional exhaustion etc.

Hence, “it can be observed that so far scholarly endeavors in this particular domain have placed disproportionately greater weight on subjects that parallel Safety I thinking, where the focus is on impairment states (and their possible links to safety violating behaviors) rather than on motivational states like one’s engagement and job satisfaction” [* which they align more to Safety-II thinking. I disagree with this line of thinking, but, whatever].

In their view, much of the existing emphasis in this domain of research has been steeped in the negative frame (which they say is Safety-I thinking), and the field could benefit further from “a stronger focus on investigating which psychosocial states are most common when everything “goes right”.

That is, to turn around the attention from safety deteriorating- to safety promoting psychosocial states (i.e., Safety II thinking) so as to capture the complexity and variety of relationships between work-induced psychosocial states and safety” [* Although I disagree that the ‘negatives’ of psychosocial factors is Safety-I, I still agree with a greater focus on the positive frames and capabilities/resources to enhance psychosocial health].

Also, it has also been argued that the existing models are “overly simplistic as they often fail to account for plausible interaction effects between several negative elements such as emotional and physical job stress, and life stressors”.

A number of limitations were observed across the research. E.g. cross-sectional, self-reported, and from an individual-level and didn’t generally reflect on the interactions across levels. Here they note that “some of the applied psychosocial constructs could be conceptualized as individual level variables (e.g., perceived role clarity), others could possibly be better off when defined as team or department level variables (e.g., psychological safety in the team)”.

This same limitation applies to the study of safety output variables. For instance, “safety output could be seen as an individual level phenomenon (e.g., individual safety compliance), a team level phenomenon (e.g., safety climate), or even an installation level phenomenon (e.g., gas leakage and explosion at an offshore oil rig)”.

Link in comments.

Authors: Derdowski, L. A., & Mathisen, G. E. (2023).?Safety science,?157, 105948.