Work-related stress assessment in an emergency department in French Guiana

Work-related stress is an issue that has a great impact on the ability of the healthcare workforce to deliver high-quality care (Basu et al, 2017). Staff in emergency departments (EDs) constantly face and have to deal with stressful situations which necessitate them adopting permanent coping strategies (Portero de la Cruz et al, 2020). Compared with staff in other departments, those working in ED are generally perceived as having a high level of autonomy, a strong sense of belonging, who are highly qualified, and equipped with good communications skills (Johnston et al, 2016). Work-related stress leads to anxiety, depression and burnout, with the result that staff who are constantly exposed to stress at work may develop a negative perception of their work. It is therefore important to continually monitor stress at work to prevent situations becoming untenable, with consequent adverse workplace effects, such as absence from work due to work-related stress, reduced productivity, and high staff turnover and attrition rates (European Agency for Safety and Health at Work, 2014; Health and Safety Executive, 2020).

There are many stressors in hospital environments, which can arise from an imbalance between work expectations and reality. A patient arriving in hospital has a certain level of expectation of the care they will receive. The health professional who sees the patient wishes to provide care to a level consistent with what they have themselves been trained to do. However, both patient and health professional may experience frustration insofar as the service provided may not always match the expectations of the care received. To ensure that the service delivered brings a high level of satisfaction, the quality of care provided should correspond as closely as possible to what is expected by the patient and what the health professional wishes to deliver.

In this context, assessing the degree of satisfaction or measuring the stressors represent two ways of examining the same problem. The first focuses on the experience of the health workers, while the second focuses on the causes of the stress. Taking such a dual approach can help identify the impact that work-related stress is having on the delivery of care and enable the development of a strategy to resolve the problem factors identified.

A public health emergency such as the COVID-19 pandemic is a major event that might heighten emotions at work and will have a major impact on work performance. Consequently, work-related stress management is an important issue for management to take into consideration. Among the many scoring systems and tools that can be used to assess levels of stress to predict burnout in the workplace (Basu et al, 2017; Cruz et al, 2019) is the Karasek model, which is considered to be the gold standard for psychological assessment in the workplace (Niedhammer et al, 2006; 2020).

Karasek assessment

The Karasek model, which has been used worldwide for about three decades, is seen as the gold standard for psychosocial jobs assessment (Kristensen, 1995). It is a three-dimensional instrument that measures decision latitude (or autonomy and skills at work), work requirements (the psychological demands of the job), and social support (from the management team and colleagues) (Box 1). The model enables a link to be made between the experiences of employees at work and the risks that their work poses to their health. The notion of job strain describes ‘the combination of high levels of psychological demands and low levels of decision latitude’ (Niedhammer et al, 2008) and is defined by medians of psychological demand and decision latitude. Another score used as part of the Karasek model is iso-strain, defined as job strain plus low levels of social support in the workplace, leading to isolation in the work environment.

Box 1.Overview of the Karasek model and its componentsThe Karasek model assesses the stress experienced by workers with regard to their physical and mental health, and is used in professional settings to assess workplace stress. The instrument aims to measure and analyse the effects of the following three factors:

• Psychological demand
• Decisional latitude, or staff autonomy
• Social support at work

The model uses data collected through a self-administered survey completed by employees. It includes 26 items, the responses to which are weighted to calculate scores to describe stress levels

Sources: European Agency for Safety and Health at Work, 2000; Expertise Collective, 2011

The Karasek model has been validated in several large epidemiological studies (Kristensen, 1995); the instrument includes the Job Content Questionnaire (JCQ), which has been translated into 22 languages (Karasek et al, 1998). It is a self-administered health and wellbeing survey that aims to measure employees' psychological and social job characteristics. One of the studies that validated the instrument is the SUMER, which is a health survey of employees' exposure to occupational risks, and was the first national French survey of its kind. It used the 26-item French version of the JCQ to obtain an overview of the stress levels experienced by employees in different socio-professional categories (Niedhammer et al, 2006).

For the study reported in this article, the Karasek model was chosen as an appropriate epidemiological tool to obtain data about the stress factors that impact on staff working in ED. However, analysis of the data collected and subsequent tests is a complex process and it may not be straightforward for team managers to transfer the results of the analysis to their own practice. In addition, a good knowledge of psychology is required by those evaluating the results to ensure robust analysis.

The Karasek model does not pinpoint the root causes of a problem and how these impact on the resulting stress-related factors experienced by staff. However, it is possible for managers to make use of the data analysis, in that it can enable them to decide which factors to focus on to pre-empt a crisis situation, because the information gained from the analysis can help them to anticipate problems and act quickly.

Aim

The objective of the present study was to assess work-related stress in the ED during the COVID-19 pandemic and the possible adverse impact on the health of staff.

Method

The study was a monocentric investigation based on a questionnaire survey administered at the onset of the epidemic. Data collection was undertaken within a short period, in April 2020, from the date that the first case of COVID-19 was admitted to Cayenne General Hospital ED. The 510-bed health facility is the referral hospital and the frontline health service for an urban population of 150 000 inhabitants and, more broadly, it is the sole provider of critical care across French Guiana, a French territory, as well as the adjacent regions of other countries located on the border with Brazil and Suriname. Cayenne is the regional capital of French Guiana and is located on the north Atlantic coast of South America. It covers 83 534 square kilometres and in 2017 the estimated population was 274 000.

The ED consists of a call centre, an out-of-hospital intervention team, an emergency unit and a short-stay observation unit, where patients may be monitored for up to 48 hours. The ED's annual throughput is more than 52 000 patients, with an average of 2500 patients hospitalised in the short-stay area. Furthermore, over one year the ED performs an average of 2300 medical interventions in the field, including 245 international medical evacuations to the West Indies or to mainland France. The team includes 40 resuscitation and emergency doctors, 180 nurses and 20 administrative staff.

The study used the French version of the Karasek questionnaire to assess work-related stress (Niedhammer et al, 2020). Karasek applies fairly simple modelling of the work environment and measures, on the one hand, constraints and, on the other, the resources needed to balance the work experience of staff against stress-related factors that pose a risk to the wellbeing of nurses and doctors and therefore the care they deliver.

Data collection

COVID-19 was considered to be a risk to all ED staff (though to different degrees), so the entire staff cohort was included in the study. All the information collected was confidential and anonymised.

The first part of the questionnaire collected epidemiological data: primary work location, professional group, age, sex, length of service in the unit, perception of workload, and conditions of work (Table 1). The second part consisted of Karasek's model and contained 24 affirmative sentences with a choice of four responses: fully agree, partially agree, partially disagree, and fully disagree.

Statistical analysis

The results were reported as median and interquartile ranges (IQR: 1st-3rd quartiles), or numbers with percentages.

Bivariate statistical comparisons were conducted using the chi-square or Fisher's exact test for categorical data and the independent t-test for continuous data. A P value of ≤0.05 was considered statistically significant. The receiver operating characteristic (ROC) curve was used to evaluate the diagnostic value of the quantitative test. The area under the curve was estimated using the Hanley and McNeill method (1982). All statistical analyses were carried out using Microsoft Excel v24.

Findings

Just over half the team, or 138 (57.5%), returned the self-administered questionnaire. Of these, 117 questionnaires were fully completed and included in the analysis.

The median age of respondents was 34 years (IQR: 29-42 years); 44.6% were male. Average length of service across all the staff in the unit was 2 years (IQR: 1-6 years). The decision latitude score was 70 (IQR: 64-74) and the psychological demand score was 25 (IQR: 23-27).

The social support by management score was 11 (IQR: 9-12) and that by colleagues was 12 (IQR: 10-12). Overall, job strain was identified in 29 respondents (24.8%) and iso-strain in 16 (13.7%).

The percentage of the total number of staff who experienced job strain was found to be 24.8%. This was higher among staff working in the call centre, breaking down as follows: job strain versus no job strain identified in 27.6% versus 6.8% (P=0.003); among the out-of-hospital team the proportions were 41.4% versus 21.6% respectively (P=0.036). Work strain was also higher among those reporting poor working conditions, identified in 24.1% versus 9.1% (P=0.036) (Table 1); it was also higher among those who reported excessive workloads (62.1% versus 23.9%; P=0.000) (Table 1).

Among nurses, the percentage was lower: job strain was identified in 31% versus no job strain in 58% (P=0.012).

The average length of service among nurses was 9 years (IQR: 3-10) in the job strain group versus 3 years (IQR: 1-6) in the no job strain group (P=0.198).

Across all respondents the percentage with iso-strain was found to be 13.7%. It was found to be higher among staff who had been in their jobs for the longest time. Among staff from the out-of-hospital team iso-strain the percentage was 56.3% versus 21.8% no iso-strain (P=0.004). Among those who thought that their work conditions were poor the percentages for iso-strain were 31.2% versus 9.9% (P=0.018). Among those who reported excessive workloads the percentages broke down as follows: 68.7% iso-strain versus 27.7% no iso-strain (P=0.001). Table 1 presents the epidemiological data of the staff members interviewed and Table 2 shows the Karasek model parameter scores for participants.

Discussion

Stress mechanisms

Among health professionals an increase in absence rate during the epidemic period is multifactorial. Reasons for absence include moral and physical exhaustion, exposure to patients with COVID-19 and consequent risk of infection with coronavirus, fear that the work environment may be unsafe and, due to potential exposure to the virus, the fear of their friends and family that they might pass on the infection, who then put pressure on the person to not go to work. These pressures can all contribute to health professionals taking the a decision to quit their jobs (Phua et al, 2020). It is therefore essential that managers do not neglect to consider the psychological impact that the current crisis is having on their workforce. However, in French healthcare facilities managers on the whole are not familiar with work-related stress management and, consequently, often neglect to deal with difficult situations at an early stage. Consequently, things will often reach breaking point before preventive action has been taken.

The promotion by management of a supportive and empathetic work culture, as part of a wider stress management programme, will help prevent situations that lead to burnout and, consequently, resignations within healthcare organisations. This should be considered an essential aspect of a good management environment in healthcare facilities. The goal must be to foster the wellbeing of all team members and not focus solely on individuals. A culture of wellbeing should be implemented for all health professionals to help reduce work-related stress. This should include assertiveness training for staff, conflict resolution management training and educating staff on how to improve their communication skills.

The authors are currently working on implementing a wellbeing programme in the ED. In this context, the study constitutes the first step towards getting an accurate measure of the gap between the situation as it currently stands and our expectations of what it should be.

Exposure to stress is an acknowledged issue in healthcare environments, with burnout situations widely described in the literature. Work-related exhaustion is the final stage of a long process of deterioration in the mental health of a health professional, with managers, unfortunately, often failing to pick this up, thereby missing opportunities for timely interventions. Interpersonal conflict, poor performance, staff turnover or resignations are logical consequences of potential neglect or failures of management policy or team leaders to address factors that can lead to staff burnout and invest in initiatives to resolve and prevent it occurring. It seems paradoxical that, although there is great awareness in the field of health care of the social and professional implications of exposure to stress at work, healthcare organisations have, on the whole, been slow to implement preventive strategies. Even though there has been a trend of adopting industrial methods in hospital management, management has generally failed to identify and address adequately the root causes of stress in the workplace.

The literature provides many examples of stress assessments. However, these instruments more often provide very generalised descriptors of stress levels and do not reveal precise information about the stressors found within a given situation. Thus, it is difficult to identify the factors that warrant the implementation of a specific plan to resolve each situation at either individual level, socio-professional group level or care unit level.

In the study described in this article, the Karasek gold-standard for assessment was used, with the resulting findings similar to those reported in the seminal 2003 French SUMER reference study. (The SUMER is a periodic national survey to describe occupation health risks.) The 2003 study collected data from a sample of 24 486 men and women of the working population. It included a self-administered questionnaire including the JCQ, the results of which were then analysed using the Karasek model (Niedhammer et al, 2006; 2020). However, good work-related stress management requires the implementation of user-friendly tools to assess the root causes of stressful situations and the implementation of institutional policies to prevent potential risk situations arising, rather than trying to fix problems once they have occurred.

Study findings

About 50% of the total staff number participated in the study reported in this article. Signs of job strain were revealed in 25% and signs of iso-strain in 14% of staff. Signs of job strain were more frequent in workers in the call centre, the out-of-hospital team, among respondents reporting poor work conditions and among those reporting excessive workloads.

Signs of iso-strain were found mainly in participants who had been in their jobs the longest, the out-of-hospital team, respondents who reported poor work conditions and those who reported excessive workloads.

In the French SUMER study, the percentage of job strain across the entire working population was identified as 23.2% and that of iso-strain as 14.5% (Niedhammer et al, 2020). The thresholds obtained were 21 for psychological demand, 70 for decision latitude, and 24 for social support. These thresholds represent the median values obtained from the calculation of the different items of the Karasek score. In the study reported in this article, the thresholds were found to be 70 for decision latitude, 25 for psychological demand, and 23 for social support.

Compared with the SUMER study of French workers, these findings were similar to the scores for decision latitude and social support, but higher for psychological demand. Nonetheless, in the reported study, the overall percentage of job strain was found to be 24.8% and that of iso-strain 13.7%—once again these findings are similar to those found in the SUMER study (Niedhammer et al, 2006; 2020). Despite the period studied being early in the COVID-19 outbreak, the ED staff showed an acceptable level of symptoms of job-strain and iso-strain. In the view of the authors, stress levels were not elevated because the two department heads and the paramedic staff have robust management training; in addition, the need to ensure psychological management of the teams was taken into consideration very early on. The teams have developed their own stress assessment and management tools and carried out iterative checks to adapt their stress management policy. This has contributed to reducing stressors and preventing emotional distress and adverse impacts on the mental health of staff. Several studies have shown that low decision latitude, high psychological demands, low social support and job strain are associated with anxiety, depression and burnout (Madsen et al, 2017; Niedhammer et al, 2020).

Limitations

The study had several limitations. First, the small sample represented about half the staff members in the unit. Second, it was conducted during a period of a health emergency, which risks overestimating the job strain and iso-strain levels. Third, the small sample size did not allow the researchers to perform subgroup analyses. However, this is the first study that assesses the stress of staff working in an ED in the Amazonian region and the French Territories of the Americas. There are plans to undertake a multi-centre study, which will involve health workers employed in EDs, intensive care units and other hospital wards. The forthcoming research will provide more details on the psychological behaviours of health workers in the French hospitals of the region.

Conclusion

It is clear that there is a rationale for implementing work-related stress management in EDs, particularly in the current circumstances of the COVID-19 outbreak. Although this study identified a level of stress similar to that reported in non-outbreak situations, it is imperative to implement stress control programmes that will help managers to anticipate and act early to mitigate the factors that pose the greatest risk for health workers developing work-related stress. Simple management tools that can be used to measure and address the work-related causes of stress, as well as the development and introduction of a wellbeing programme for health professionals working in the ED, are required.

KEY POINTS

• The study has emphasised the importance of undertaking a work-related stress assessment to identify pressures on staff, to enable managers to implement measures to pre-empt a crisis situation
• Having a stress management policy in place will prevent situations at work reaching a point that leads to staff burnout
• A supportive management culture is vital for preventing staff burnout and high staff turnover
• Monitoring stress at work can prevent situations of staff burnout, the consequent adverse of which can be staff absences, reduced productivity, and high turnover and attrition rates
• The use of a stress assessment tool can enable information to be gained and analysed, helping management to anticipate problems and act quickly

CPD reflective questions

• Consider the situation in your workplace. Do you consider that stress at work is inevitable?
• If you are a manager, do you think that you are managing the stress of your teams well? If you are a more junior staff member, how well do you think your managers are managing stress within you team?
• Do you think that you, or your managers, have the necessary tools to provide supportive and compassionate management?
• Consider your organisation's stress-management policy. Do you think it is enables stress to be managed well?