Elsevier

Safety Science

Volume 93, March 2017, Pages 143-151
Safety Science

A method of identifying health-based benchmarks for psychosocial risks at work: A tool for risk assessment

https://doi.org/10.1016/j.ssci.2016.11.016Get rights and content

Highlights

  • Psychosocial risks are well-recognised occupational health hazards.

  • No critical exposure levels exist for psychosocial risks to inform risk management.

  • We offer a method to estimate critical levels of exposure for psychosocial risks.

  • We identify critical levels of job control using the benchmark dose method.

  • Influence over many aspects of work is needed to reduce the risk of depression.

Abstract

Objectives

We present a novel approach to identify critical exposure levels or health-based benchmarks of job control using the benchmark dose (BMD) method. This method provides benchmarks for risk assessment of psychosocial risks, similar to benchmarks used for other occupational health hazards such as chemicals.

Methods

Two staged (bivariate and adjusted) BMD modelling was conducted using epidemiological data from an age-cohort study in south-eastern Australia. The adjusted BMD model incorporated age, gender, education, personality traits and mental health status at baseline.

Results

Depression is a more sensitive (health compromising) outcome for job control compared to anxiety in both types of BMD modelling. For an excess risk of 5% for depression, the adjusted benchmark dose was 0.49 and the critical exposure level, being the lower one sided 95% confidence limit of the adjusted BMD, was 0.37. If workplace guidelines are based on this critical exposure level, workers need to have a minimum of ten out of 15 aspects of job control measured in this study to reduce the excess risk of depression.

Conclusions

The BMD approach can identify critical exposure levels for job control. This suggests a similar approach can be used for other psychosocial risks for which no critical exposure levels are currently available. Critical exposure levels can provide guidance needed to assess risk and address psychosocial risks, similar to other health hazards. Benchmarks or critical exposure levels of psychosocial risks can assist the inspectorate and employers to conduct risk assessment of workplaces and identify areas for intervention.

Introduction

Psychosocial risks which can arise from the organisation and design of work are associated with chronic health conditions in workers including depression, anxiety, cardiovascular disease and musculoskeletal disorder in both affluent and transitional economies (Backé et al., 2012, Clumeck et al., 2009, Stansfeld and Candy, 2006, Sultan-Taïeb et al., 2011). In Australia, workers’ compensation claims for psychosocial risks are among the most common and the most costly type of occupational disease claims (Safe Work Australia, 2012, Safe Work Australia, 2013). A 2010 survey of occupational health and safety (OHS) experts from twenty developing countries including Namibia, Chile, India and China identified work design hazards as the second highest priority for OHS in developing countries after accident prevention (Kortum et al., 2010).

Different theoretical models of psychosocial risks and its effects exist and one of the most well-known is Karasek’s demand-control model (Karasek, 1979, Stansfeld and Candy, 2006). The demand-control model identifies two psychosocial risks: high job demands and low job control. Job control is control over the type and timing of work tasks and skill discretion that enables a worker to decide which skill to use and how to organise and complete their tasks. This is especially important in coping with job demands which include heavy workloads, fast pace of work, and deadlines.

Common mental disorders such as depression and anxiety are the outcomes with the most epidemiological evidence available in relation to low job control (Clumeck et al., 2009, Stansfeld and Candy, 2006, Stansfeld et al., 2008). These associations appear to be universal. A study of Thai workers shows that job-related risk factors linked to time pressure and a lack of decision making authority (termed job control) were associated with psychological distress even after adjusting for baseline mental distress and exposure to physical hazards (Yiengprugsawan et al., 2015). Longitudinal studies in Japan, France, UK, Australia and Taiwan also show that low job control significantly predict depression and anxiety in workers after controlling for potential confounders (Butterworth et al., 2012, Huang et al., 2012, Inoue et al., 2010, Niedhammer et al., 1998, Stansfeld et al., 2008, Strazdins et al., 2011).

Despite the strong empirical evidence, there is the perception that psychosocial risks such as job control are too complex and difficult to manage, and that they are rarely systematically assessed or managed in the workplace (Iavicoli et al., 2011). In the first EU wide survey of businesses on psychosocial risks in 2009, only half of the businesses reported that they had informed their employees about psychosocial risks and 42% considered it more difficult to tackle psychosocial risks compared to other health and safety hazards (European Agency for Safety and Health at Work, 2010). The most common reasons for this perceived difficulty were the lack of awareness and lack of resources. A second survey in 2014 showed that employers still had difficulty in addressing psychosocial risks with a common reason being a lack of adequate tools to deal with the risk effectively (European Agency for Safety and Health at Work, 2015).

There is also continued resistance to addressing psychosocial risks in the workplace by some workplace stakeholders and conflicting views among various stakeholders about responsibility (Gordon et al., 2009, Leka et al., 2015). The very nature of psychosocial risks arising from the organisation of work means that improving the psychosocial work environment will require changes in how employers manage and run their workplaces. Addressing psychosocial risk factors for poor mental health also poses additional challenges due to the complexity in measurement and assessment of causality for mental disorders.

Part of the challenge in addressing psychosocial risks is the way these risks are legislated and the type of information currently available to assist employers. In Australia, a federation with nine main OHS legislations (federal and eight states and territories), general duty requirements in occupational health and safety legislations cover psychosocial risks, where health is defined to include psychological health in most jurisdictions. However, as psychosocial risks are not explicitly defined, employers are required to recognise the range of psychosocial risks that can harm workers’ health and to be aware that their duty to provide a healthy and safe environment to workers extends to psychosocial risks in the workplace. While a number of codes of practice, guidance materials and factsheets are produced by various Australian OHS regulators and nationally by Safe Work Australia to help address psychosocial risks, none include specific benchmarks or exposure levels to guide employers to determine when action or intervention is required in their workplace to protect workers’ mental health.

This difficulty in determining a safe and healthy psychosocial work environment is an issue not just limited to employers but also for OHS inspectors who risk assess, monitor and enforce health and safety laws. In a study of OHS inspectors in Australia, only 5% of issues identified during workplace visits related to psychosocial risks, with the majority of inspections concentrating on ‘traditional’ occupational hazards such as machine guarding and noise (Johnstone et al., 2011). Even when psychosocial risks were addressed, they were largely limited to the more tangible issues such as bullying and harassment complaints rather than the underlying psychosocial risks arising from the organisation of work such as job control and job demands at work. This is partly because OHS inspectors are used to focusing on physical hazards such as machine guarding, noise and hazardous chemicals which can be inspected using a check-list, a walkthrough of the workplace or by undertaking noise and airborne measurements in the workplace. In contrast, psychosocial risks are typically not visible and unlike noise and chemicals, there is no voluntary or mandatory guidance on which levels of exposure to psychosocial risks are acceptable. This makes the determination of what constitutes a safe psychosocial work environment difficult and specialist knowledge and expertise is required to do so.

OHS inspectors in other countries also face similar issues and have developed and trialled new methods for inspection of psychosocial risks (Ertel et al., 2010). Bruhn and Frick (2011) highlighted the challenges faced in Sweden when developing specific methods and tools for psychosocial risks that can be used by inspectors without special training or background in psychosocial risks. Despite development of new methods and tools, Bruhn and Frick concluded that in practice, these tools provided little guidance, especially for non-specialist inspectors without a background in psychological or social sciences. In Denmark, the Danish Working Environment Authority (DWEA) has developed industry sector specific guidance tools and trained the labour inspectors in how to use these guidance tools to assess and evaluate psychosocial risks (Rasmussen et al., 2011). However, the use of the tools was considered challenging for inspectors lacking a background in psychology and inspectors were still required to make an individual evaluation of whether a workplace’s psychosocial work environment requires improvement and make judgements in relations to health risks.

Limited knowledge on how to assess psychosocial risks and general regulations lacking in clarity were also identified as barriers to reducing exposure to psychosocial risks in other European studies (Graversgaard, 2004, Iavicoli et al., 2011, Leka et al., 2011). An investigation of labour inspection strategies in European Union Member States for emerging risks such as psychosocial risks found that the lack of visibility of psychosocial risks, the multi-causal nature of mental disorders and the reluctance by individuals to report issues associated with psychosocial risks further add to this challenge (Cardiff University, 2011). The authors concluded that support for risk assessment and risk management of psychosocial risks is needed to improve labour inspection responses (Cardiff University, 2011).

In recent years, there have been many advances to improve risk assessment and risk management of psychosocial risks. Examples include the UK Health and Safety Executive (HSE) Management Standards and the indicator tool, the European Psychosocial Risk Management Framework (PRIMA-EF), a standard for psychosocial risk management (PAS 1010) which provides guidance on managing psychosocial risks systematically (British Standards Institution, 2011, Cousins et al., 2004, Leka et al., 2008) and a voluntary Canadian standard for psychological health and safety in the workplace (Canadian Standards Association, 2013). Recognising the variation and difficulties in undertaking inspection of psychosocial risks at work, the International Labour Organisation is also developing guidelines for labour inspection of psychosocial risks to provide global guidance on this issue but the tool is not yet publically available (Velaquez, 2015).

Risk assessment tools such as the HSE’s indicator tool enable a workplace to undertake risk assessment that identifies hazards and to assess risk of potential harm associated with exposure to these hazards. Risk assessment tools for psychosocial risks are typically surveys and can be used by health and safety inspectors during workplace inspections and by managers and employers to conduct risk assessment in their organisations. For example, in Denmark, the Copenhagen psychosocial questionnaire is used by Danish companies using the national average from a population study (Kristensen et al., 2002) as a benchmark (Pejtersen et al., 2010). However, the HSE and Danish benchmarks are based on organisational or national averages or percentiles and these benchmarks do not provide an indication of health risks.

Our examination of the regulatory and risk management approaches for other occupational health hazards such as chemicals and noise suggests that critical exposure levels or health-based benchmarks for psychosocial risks may provide an alternative approach to risk assessment of psychosocial risks (National Occupational Health and Safety Commission, 2000, Nielsen and Øvrebø, 2008). Critical exposure levels are levels of exposure that are associated with adverse health outcomes. Employers and other occupational health and safety stakeholders are already familiar with this approach and it can complement existing risk management frameworks for psychosocial risks such as the PRIMA-EF and PAS 1010.

Critical exposure levels provide a way of translating evidence about occupational health hazards into prevention activities and policies. First, critical exposure levels can help in risk assessment and risk management of psychosocial risks by providing benchmarks for risk assessment based on adverse health effects. This will enable a proactive approach to prevention of adverse health outcomes associated with psychosocial risks rather than acting only when the worker becomes ill or lodges a workers’ compensation claim. Critical exposure levels can also provide a more solid base for regulatory enforcement and other inspectorate activities that may act as a stimulus for action in workplaces and may assist in the evaluation of health risks by non-specialist inspectors.

The current study seeks to apply the benchmark dose (BMD) method, a dose response modelling method for identifying critical exposures levels, to psychosocial risks using job control as an example and depression and anxiety as health outcomes. The BMD method is chosen because it can use data from epidemiological studies. This means potential confounders can be taken into account. For example, the level of job control varies by gender and socioeconomic status (Dalgarda et al., 2009, Schrijvers et al., 1998, Smith et al., 2008). Common health outcomes associated with low job control such as depression and anxiety are also usually multifactorial. Therefore, taking known confounders into account is important for establishing accurate critical exposure levels for complex hazards such as job control, and for health outcomes such as depression and anxiety.

Section snippets

Population and study design

Data were from the first two waves of the PATH Through Life project, an age-cohort study in South East Australia conducted by the Australian National University (Anstey et al., 2012). The study sample was randomly drawn from the electoral roll which contains over 94% of those who are eligible to vote. Over 58% of the enumerated sample participated in the study and the attrition from wave 1 to wave 2 ranged from 7% to 11%. There was a four year period between waves. The current study used data

Characteristics of the sample at wave 2

Table 1 shows the distribution of the sample regarding socio-demographic characteristics and the prevalence of depression and anxiety. Although representative of the region from which the sample was drawn, our sample had higher education level and fewer blue collar workers compared to the Australian population. Approximately 45% of the sample had a bachelor/postgraduate degree. The prevalences of depression and anxiety were similar between the two waves with approximately 8% and 4% with

Discussion

Despite well documented health hazards stemming from psychosocial risks, regulation and risk assessment of psychosocial risks are still lagging behind other occupational hazards and are still viewed as complex. Thus, the application of the critical exposure level approach for psychosocial risks may provide a way of improving risk assessment and regulation of these hazards. This approach can add to the existing guidance and resources available while providing a method of risk assessment that is

Conclusion

Our findings show that the BMD approach is feasible for identifying critical exposure levels for job control and poor mental health. This provides the foundation for future BMD studies of psychosocial risks. While further work on a BMD based risk assessment tool is required, the study shows how a benchmark can be used as a risk assessment tool in workplaces. The use of health-based benchmarks for psychosocial risks that take into account individual variation in vulnerability may help mainstream

Funding

Funding for the PATH study was provided by a unit grant (no. 973302) and new program grant (no. 179805) from the National Health and Medical Research Council (NHMRC) – Australia, and additional support from the Australian Rotary Health Research Fund and the Australian Brewers’ Foundation Medical Research Grant Scheme. Safe Work Australia provided the funding for the writing up of this article. However, the views expressed in this article are that of the authors and not of Safe Work Australia.

Contributors

SK wrote the analysis plan, cleaned data, conducted analyses, interpreted results and drafted and revised the manuscript. All contributed to the study design. LS made critical revisions to the manuscript. All contributed to interpretation and discussion of results. MC, LG and PB reviewed and revised drafts. MC advised on statistical analysis and conducted analyses.

Acknowledgements

The authors would like to thank the participants in the PATH study; the interviewers for the study; Patricia Jacomb and Karen Maxwell for managing the study and data collection; Tony Jorm, Helen Christensen, Kaarin Anstey, Keith Dear and Bryan Rodgers who were the chief investigators for the first two waves PATH study.

References (60)

  • E.-M. Backé et al.

    The role of psychosocial stress at work for the development of cardiovascular diseases: a systematic review

    Int. Arch. Occup. Environ. Health

    (2012)
  • L.I.V. Bergh et al.

    Developing a performance indicator for psychosocial risk in the oil and gas industry

    Saf. Sci.

    (2014)
  • British Standards Institution

    Guidance on the Management of Psychosocial Risks in the Workplace, PAS 1010:2011

    (2011)
  • E. Budtz-Jorgensen et al.

    Benchmark dose calculation from epidemiological data

    Biometrics

    (2001)
  • P. Butterworth et al.

    Common mental disorders, unemployment and psychosocial job quality: is a poor job better than no job at all?

    Psychol. Med.

    (2012)
  • Canadian Standards Association, 2013. Psychological Health and Safety in the Workplace – Prevention, Promotion, and...
  • Cardiff University, 2011. Contract to Assess the Potential Impact of Emerging Trends and Risks on Labour Inspection...
  • C.S. Carver et al.

    Behavioral inhibition, behavioral activation, and affective responses to impending reward and punishment: the BIS/BAS scales

    J. Pers. Soc. Psychol.

    (1994)
  • N. Clumeck et al.

    Working conditions predict incidence of long-term spells of sick leave due to depression: results from the Belstress I prospective study

    J. Epidemiol. Community Health

    (2009)
  • R. Cousins et al.

    Management Standards work-related stress in the UK: practical development

    Work Stress

    (2004)
  • K. Crump et al.

    Benchmark concentrations for methylmercury obtained from the Seychelles Child Development Study

    Environ. Health Perspect.

    (2000)
  • O.S. Dalgarda et al.

    Job demands, job control, and mental health in an 11-year follow-up study: normal and reversed relationships

    Work Stress

    (2009)
  • J.A. Davis et al.

    Introduction to benchmark dose methods and U.S. EPA’s benchmark dose software (BMDS) version 2.1.1

    Toxicol. Appl. Pharmacol.

    (2010)
  • M. Ertel et al.

    European social dialogue on psychosocial risks at work: benefits and challenges

    Eur. J. Ind. Relat.

    (2010)
  • European Agency for Safety and Health at Work

    European Survey of Enterprises on New and Emerging Risks: Managing Health and Safety at Work, European Risk Observatory Report

    (2010)
  • European Agency for Safety and Health at Work

    Second European Survey of Enterprises on New and Emerging Risks: Managing Safety and Health at Work, European Risk Observatory Report

    (2015)
  • J. Fuchs et al.

    Genotoxic risk for humans due to work place exposure to ethylene oxide: remarkable individual differences in susceptibility

    Arch. Toxicol.

    (1994)
  • D. Goldberg et al.

    Detecting anxiety and depression in general medical settings

    Br. Med. J.

    (1988)
  • D.R. Gordon et al.

    Stakeholder perspectives on work and stress: seeking common ground

  • J. Graversgaard

    Key role of labour inspection: how to inspect psychosocial problems in the workplace?

  • Cited by (15)

    • Physical hazards, musculoskeletal pain, and job control: Multilevel results from the European Social Survey

      2022, Safety Science
      Citation Excerpt :

      Among different moderators, job control stands out as a possible buffer. Job control is defined as the autonomy employees have over deciding their tasks and work schedule (Kyaw-Myin et al., 2017). Job control is theorized as a type of job resource (e.g., job demands and resources theory, Sorlie et al., 2022).

    • A comparison of risk measures for accidents in the energy sector and their implications on decision-making strategies

      2018, Energy
      Citation Excerpt :

      The consideration of economic cost aspects in accidental events is also a clear focus in the insurance industry that traditionally addresses different types of risk and their potential impacts on infrastructure facilities and the services they provide by looking at historical data and loss modelling [31,32], which nowadays is also complemented by predictive modelling, optimization approaches, etc. to address changes in the risk landscape and to react to emerging and novel risks [33–36]. Scholtens and Boersen [37] looked at the response of stock markets to major energy accidents, whereas other studies analyze the cost of reputational damage resulting from major accidents [38] or address aspects of risk aversion, perception and beliefs [39,40]. Finally, the well-known approaches for stress testing applied in the financial industry have been adapted to provide stress tests for critical infrastructures including cascading effects and multi-hazard scenarios [41].

    View all citing articles on Scopus
    View full text