Abstract
Introduction
Intensive care staff have high levels of stress. We conducted a service improvement initiative to assess workplace stress levels among staff in one adult general intensive care unit and deliver a stress management intervention.
Methods
A psychological intervention of four stress management sessions, and fortnightly staff support drop-in groups, was developed and delivered within a year. Pre- and post-intervention, workplace stress in the unit was assessed using a Health and Safety Executive tool.
Results
Pre-intervention assessment of 76 (47.2%) staff indicated that improvement was needed in all domains of workplace stress. 125 staff (77.6%) participated in the intervention and gave positive ratings for content, relevance, practicality and personal value (median 4 (1–5); interquartile range 3.8–4.6). Post-intervention assessment of 71 staff (41.3%) demonstrated improvements in all workplace stress domains.
Conclusion
A reduction in workplace stress was observed following a service improvement intervention in one intensive care unit although no causality can be assumed. Similar interventions should be evaluated using robust study designs.
Keywords: Workplace stress, burnout, stress management intervention, intensive care staff
Introduction
Intensive care staff experience high levels of stress, affecting well-being and performance at work. Excessive workplace stress has been linked to staff experiencing anxiety, depression or burnout (defined as emotional exhaustion, cynicism and inefficacy).1–5 These difficulties may also contribute to rapid staff turnover, a constant challenge for intensive care units (ICUs).6 Critical care societies in the US recently called for action to diminish the harmful consequences of burnout syndrome7 and the UK Intensive Care Society widened the focus to promoting workplace well-being as well as preventing burnout in ICU staff.8
Research from several countries has focused on ‘burnout’ and clinical consequences of stress such as depression. In the UK, 29% of 627 intensive care doctors were found to suffer distress and 3% had suicidal thoughts.1 In France, 46.5% of doctors from 189 ICUs and 33% of 2392 ICU nurses were suffering burnout.2,3 Similarly, 31% of Portuguese intensive care staff were found to have burnout.4 While workplace stress affects all healthcare staff, a recent report showed the highest rates of burnout, depression or both (54%) among physicians working in intensive care.5
Few interventions to reduce stress or to improve the well-being of critical care professionals have been evaluated. One single-centre study of 29 ICU nurses found that a 12-week resilience training programme was acceptable and feasible to participants.9 In another small study, an hour of yoga, mindfulness and music delivered for eight weeks increased work satisfaction among surgical ICU staff.10 A systematic review of interventions for general clinician burnout found that both individual-focused and organisational strategies worked, and hypothesised that a combination of the two elements would create the most improvement.11
Rather than focus on the most serious outcomes such as PTSD, or burnout, we aimed to assess workplace stress among our staff, including specific critical care factors, to help us identify sources of stress, and potentially prevent more serious outcomes. To develop a psychological intervention for our ICU staff, we first reviewed the literature on major stressors and risk factors for ICU workplace stress. Major stressors included: difficult working relationships, work-load, being ‘over-stretched’, compromising standards, lack of resources, lack of recognition, too much responsibility, making decisions alone and bureaucracy.1 Independent risk factors included female gender, age, organisational factors, workload, quality of working relationships, conflicts, end-of-life factors and ethical decisions relating to withdrawal of therapy.2–5
We aimed to develop a stress management intervention to address modifiable risk factors identified in the literature review, by giving staff tools to improve working relationships, reduce conflicts, and manage emotional factors and ethical decisions around end-of-life care. In addition, we planned to use the results of the workplace stress assessment, and to consult with staff on their sources of stress, to further refine the content of a stress management intervention, as well as to identify organisational stressors. The organisational stressors would be fed back to the ICU management team in parallel with the psychological intervention, as we recognised organisational factors cannot be addressed by stress management alone.
Aim
Our aim was to implement and evaluate a service improvement initiative to assess and reduce workplace stress in the University College Hospital London (UCLH) ICU. Objectives were to assess the level and sources of work-related stress among staff, identify contributing factors, develop and deliver a psychological intervention for staff based on existing evidence and the assessment, and evaluate the impact of the intervention on workplace stress.
Methods
Design
Delivery and evaluation of a staff stress management intervention, with stress assessments pre- and post-intervention. Workplace stress was measured by an anonymous questionnaire among ICU staff in March 2013. Following this, a stress management intervention was delivered to staff and evaluated. In March 2014, the workplace stress assessment was repeated, with additional questions about staff experience of the intervention.
Measures
No validated questionnaire measuring workplace stress in ICU staff could be identified. We therefore used the validated Health and Safety Executive (HSE) management standards indicator tool – a 35-item questionnaire that assesses the seven recognised factors (domains) of workplace stress: change, role, peer support, manager support, relationships, demand and control.12 To this we added a bespoke questionnaire based on existing literature, consisting of 18 items about common stressors experienced by ICU staff. The items broadly fall into three recognisable categories: institution-related (staff shortages, lack of equipment, lack of beds, lack of sleep (own), external pressures and targets); patient-related (distressed patients or relatives, agitated patients, dying patients, futile treatment) or staff-interaction-related (difficult relationships with own professional group/other professional group/non-ICU staff, fear of blame, fear of making mistakes, lack of recognition, poor communication among staff). A multiple choice question about job satisfaction and several items on demographic data (age, gender, professional role and years of employment) were also included in the questionnaire.
The questionnaires were previously piloted with a sub-section of staff, who indicated their preference for an anonymous questionnaire with minimal demographic data.
Procedure
Workplace stress assessments: Paper copies of the questionnaire were distributed by senior clinicians to all staff members. Staff could complete questionnaires in private and post them anonymously in a box. As questionnaires were completed anonymously at both time points, it is unknown whether staff completed one, both or no questionnaires. All staff received an email informing them about the assessment.
The intervention
A psychological stress management intervention was developed, aiming to help clinical staff understand their own and others’ stress, to discuss stress more openly with colleagues and to support each other in developing coping strategies (Figure 1). It also aimed to enhance communication within and between ICU teams, and to improve patient care and compassion by helping staff manage difficult emotions. The intervention comprised taught stress management sessions and staff support drop-in groups, designed and run by psychologists from the staff psychology service and the intensive care department. Different methods of education and communication, including didactic teaching, personal reflection, role plays, relaxation practice and small discussion and support groups, were used to accommodate different learning styles and emotional needs.
Figure 1.
Content and structure of the staff stress management sessions and drop-in groups.
A total of 20 taught stress management sessions were run throughout the year during nursing team days (four sessions/workshops run five times each) to give most nurses the opportunity to attend. Attendance at team days is compulsory, although some nurses might be on annual leave for at least one team day. All other critical care professionals were invited to attend the stress management sessions on a voluntary basis, if they could be released from clinical duties. The four 90 minutes sessions had the following themes: (1) Identifying signs and sources of stress at work (2) communication skills to support team work and to manage patients’/relatives’ anxieties (3) dealing with difficult emotions around serious illness and end-of-life and (4) strategies for coping with stress (such as guided imagery, relaxation exercises, cognitive techniques and mindfulness). During sessions, participants were helped to make a personal plan to deal with stress and improve coping.
The quality of each stress management session (content, relevance, presentation, practicality and personal value) was evaluated using feedback forms, and a scale to rate individual stress from 0 (no stress at all) to 100 (most stress I have felt) was used in sessions 1 and 4 to detect any change in stress levels from the beginning to end of the sessions. Fortnightly staff support drop-in groups run by psychologists in the critical care unit were open to all staff, and senior staff were requested to make arrangements for those who wanted to attend. No register of attendance at drop-in groups was kept, or feedback asked for, as attendance was voluntary and kept confidential, as requested by staff. Due to anonymity, it was not known if staff who participated in the stress management intervention also took part in either or both stress assessments. For further details of the content of the sessions and groups, see Figure 1.
Data analysis
Data are presented as means and standard deviations, or medians and interquartile range, where appropriate. Regarding the HSE management standards indicator tool, scores for each factor were calculated and benchmarked against 136 UK organisations. The results fall within colour-coded percentile ranks requiring follow-up actions: urgent action (<20th percentile, red zone); clear need for improvement (20th–50th percentile, yellow zone); good, but need for improvement (50th–80th percentile, blue zone); or performance to be maintained (>80th percentile, green zone). Survey results from 2013 and 2014 were compared in terms of percentile rank for each of the seven factors.
Governance
As this was a voluntary and anonymous staff survey, we did not seek ethics approval. The chief nurse and clinical lead of ICU were consulted, and the project was discussed and approved at the critical care governance meeting.
Results
Workplace stress assessment
Demographics
Pre-intervention, 76 members of staff (47.2% of the staff establishment of 161) took part in the workplace stress assessment. The majority of the workforce and participants constituted nursing staff and were female. Mean age of participants was 35 years and duration of employment within the critical care directorate was 5.3 years. In the 2014 post-intervention assessment, the number of participants was marginally fewer (71/170 (41.3%) of the staff establishment), and a one-third reduction in nursing participants was offset by a two-fold increase in both medical and therapist respondents. There were 15% fewer female participants in 2014, while age and duration of employment were similar at 34 and 5.6 years, respectively (Table 1).
Table 1.
Workplace stress assessment: Population demographics.
| 2013 | 2014 | |
|---|---|---|
| Professional role | ||
| Nursing, n (% of total nursing staff) | 62 (49.6%) | 47 (33.6%) |
| Medical, n (% of all medical staff) | 11 (37.9%) | 16 (69.6%) |
| Therapist, n (% of all therapists) | 3 (42.9%) | 6 (85.7%) |
| Other staff, n (% of other) | 0 | 2 (100.0%) |
| Total, n (% of all staff) | 76 (47.2% of 161) | 71 (41.8% of 170) |
| Age (years), median ± IQR | 35 ± 9 | 34 ± 11 |
| Female, n (% or responders) | 62 (81%) | 47 (66%) |
| Employment (years), median ± IQR | 5.3 ± 6.8 | 5.6 ± 7.8 |
IQR: interquartile range.
Job satisfaction
Responses to the ‘job satisfaction’ items of the assessments demonstrated a reduction from 2013 to 2014 in those who ‘did not enjoy their job’ (9.2 to 4.2%) and those that ‘generally enjoyed their job’ (78.9 to 70.4%), coupled with an increase in those who were ‘completely happy with their job’ (11.8–19.7%).
Workplace-related stress
The HSE management standards indicator tool demonstrated that workplace-related stress factors prior to the stress management intervention in 2013 were as follows: demand and control were below the 20th centile; relationships, peer and manager support were between the 20th and 50th centile; change and role were between the 50th and 80th centile. After delivery of the psychological intervention in 2014, an improvement was demonstrable in all seven domains, with change and role achieving the highest rating (see Figure 2).
Figure 2.
Graph showing positive change in all workplace stress factors from 2013 to 2014. The Health Standards Executive management standards indicator tool was used to assess work-related stress. Results are calculated as percentiles and action required after plotting against the results of 136 reference organisations.
Green > 80 centile – doing very well, need to maintain performance; Blue 50–80 centile – good, but need for improvement; Yellow 20–50 centile – clear need for improvement; Red < 20 centile – urgent action needed.
Definitions of factors: Change – how organisational change is managed; Role – whether people understood their role; Peer support – encouragement and resources from colleagues; Management support – encouragement, sponsorship, resources provided by organisation/line management; Relationships – promoting positive working to avoid conflict and deal with unacceptable behaviour; Demand – workload, work patterns and environment; Control – how much say a person has in the way they work.
Critical care stress factors
The ratings for specific critical care stressors reported by different critical care staff groups are shown in Figure 3.
Figure 3.
Top intensive care stressors – by staff group. The top 5 intensive care stressors for each staff group from the workplace stress assessment in 2013, which included 18 items on common stressors in critical care.
Orange: institution-related factors; yellow: patient-related factors; and grey: staff interaction-related factors.
Most commonly reported institution-related factors were staff shortages, lack of equipment, lack of beds (for admissions) and lack of sleep (medical staff). Most reported patient-related factors were caring for agitated, delirious or distressed patients; and futility of treatment. Staff interaction-related factors were fear of making mistakes or being blamed, lack of recognition, poor communication between clinicians (medical staff and/or medical staff and wider multi-disciplinary team) and difficult working relationships.
Intervention evaluation and impact of stress
125 nursing staff (90% of nursing workforce; 77.6% of total workforce) attended four stress management sessions/workshops each. No doctors or allied health professionals attended the sessions. Each session was evaluated by attendees for its content, relevance, presentation, practicality and personal value; median score 4 (IQR 3.8–4.6; 1 = strongly disagree, 5 = strongly agree). Stress ratings (0–100) were measured at the beginning and the end of the first and the fourth workshops. A fall in stress levels was seen during session 1, from a median rating of 40 (IQR 20–60) at the beginning to a median of 20 at the end of the session (IQR 0–30). During session 4, a stress reduction was also reported, from median 50 (IQR 30–75) to median 30 (IQR 10–40).
In the 2014 workplace stress assessment, 61% participants reported they had attended the stress management sessions, and 70% said they wanted the critical care unit to continue running the sessions. The vast majority wanted to maintain the initial format of stress management sessions during team days, and staff support drop-in groups during the working day, with only 15% disliking this format.
Discussion
We identified the need and carried out a project to assess and manage workplace stress among our intensive care workforce. Similar to other surveys carried out in intensive care,1–4 our first assessment highlighted worrying levels of staff workplace stress. A staff stress management intervention, designed and run by psychologists embedded in the ICU and from the staff psychology service at UCLH, was feasible to deliver in practice and acceptable to staff. Following the intervention, a reduction in all seven domains of workplace stress and an increase in job satisfaction were observed, although no causality can be assumed. Throughout the project, progress and information on staff stressors was fed back to all senior staff and the management team of the ICU.
The strength of our approach was that we used the HSE management standards indicator tool as a validated measure of workplace stress, which is able to indicate both the sources as well as the levels of stress. Previously published work focused on measuring clinical outcomes such as anxiety or depression, which can both be affected by external factors that are not related to the workplace. We also identified specific sources of intensive care stress for different staff groups with a bespoke questionnaire. The main sources of intensive care stress for nurses and therapists were organisational – staff shortages – while the main stress for doctors was fear of making mistakes, and for nursing assistants caring for agitated, delirious patients, a responsibility that few would have been trained in. Generally the main stress could be attributed less to patient factors and more to organisational factors which also included lack of equipment, and to staff interactions such as fear of being blamed for mistakes, lack of recognition (assistants, therapists), poor communication and difficult working relationships. In addition, our aim was to intervene and alleviate staff stress, with the delivery of a stress management intervention designed specifically for staff in the intensive care setting. The intervention was attended by 90% of nurses or 77.6% of the total ICU staff.
Limitations were that this was a service improvement initiative conducted in a single centre, with a pre-post design and no control group for comparison. Although the reduction in workplace stress in seven domains was encouraging, many external factors in addition to the stress management intervention could have influenced this result. Workplace stress assessments were conducted with 47.3% and 41.7% of eligible staff at both time points. There was a risk of bias in the survey, in that staff with more stress might have been more motivated to complete the questionnaire. As the assessments were anonymous, and there was staff turnover between 2013 and 2014, scores did not come from exactly the same groups of staff at both time points. Barriers to implementing the stress management intervention included staff shortages, meaning that doctors and allied health professionals did not attend the workshops (although some attended drop-in groups); and an impression that some staff were not comfortable to share their emotions with colleagues, particularly those from a different healthcare profession.
The taught stress management sessions were designed to help staff to modify their reactions, to reduce the impact of stressors on them and to help them develop new coping skills. During the first stress management session, staff described their strategies for coping with stress at work as taking practical concrete steps to solve problems, seeking social support and doing pleasurable activities in leisure-time. This mainly problem-focused coping style does not help staff cope with situations at work that are not remedial to practical solutions, such as dealing with distressed relatives or with their own feelings about end-of-life care. To address this, we provided opportunities (in sessions 2 and 3) for small groups of staff to express their feelings and beliefs about coping with death and dying, as well as addressing patients’ or relatives’ anxieties. Based on evidence from a Cochrane Review13 that cognitive behavioural therapy (CBT) and mental and physical relaxation reduce stress, and a request from staff for practical strategies to cope with stress, a range of CBT and relaxation tools to alleviate stress were taught in the final session.
Key factors influence the effectiveness of stress management interventions.14 First, staff involvement in the design increases engagement in the intervention. In the first stress management session, staff were asked about factors that they believed contributed to their work-related stress and informed that these factors would be addressed in a tailored stress management intervention specifically for them. Second, managerial support for stress management interventions is a key factor.15 We were fortunate that there was demonstrable senior management commitment to the stress intervention in our unit. Time was made available for stress management sessions in the nurses’ education programme on team days, and staff were informed they could be relieved from clinical duties to attend drop-in groups.
A third key factor in effectiveness is the organisational climate (i.e. what it is like to work in a specific setting such as healthcare, specifically critical care). A strong culture exists in healthcare of being task-focused and having a stoic attitude as a defence mechanism. There was a mixed response to small group discussions about staff members’ emotions about dying patients or anxious families. Some staff did not find it helpful to focus on their feelings, while others were pleased to know they were not the only ones to be affected by loss or other painful emotions at work. On occasions, staff were concerned that attendance would be perceived as weak or resulting from an inability to cope. Views were expressed during feedback such as: ‘If you cannot cope with this sort of work, then maybe it’s just not for you’ or ‘going to the drop-in sessions makes me feel I am trying to get more break time’. Others gave positive feedback: ‘The session was very applicable to the workplace’ and ‘it’s about time we talked about how work affects us’. The intervention, as well as feedback to ICU senior staff/management about organisational factors, may have contributed to developing a more supportive management style and increased peer support for each other. At the end of the initiative, intensive care managers decided to continue the stress management intervention as a permanent service for staff, due both to the reduction in stress seen and the finding that 70% of respondents told us they would like it to continue.
Conclusion
A heavy workload, together with sub-optimal support from peers and managers, and a lack of personal control, were major sources of stress for staff in one ICU. A reduction in workplace stress was observed following the implementation of a stress management intervention, although no link can be implied. More work should be done to develop multi-disciplinary and multi-modal interventions to reduce critical care staff stress and to evaluate them in randomised trials.
Acknowledgements
We would like to thank Nicola Gale for suggesting use of the HSE indicator tool and helping to devise the stress management sessions, and Anthony Hazzard, Father Peter Harries and Jo Bennetts for contributing to the content of the sessions and helping to run them. Thanks to Elaine Thorpe and David Howell for their management support of the workplace stress initiative.
Footnotes
Authors’ contributions: HG, MG, NM and DW conceived this service improvement initiative. MG developed the stress assessment questionnaire with help from DW. MG and HG organised the staff surveys. JF led on developing the stress management sessions with DW. HG and DW led on delivery of the stress management programme. DW, MG and NM drafted the manuscript, and other authors critically reviewed the manuscript.
Declaration of conflicting interests: The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding: The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This project was funded by Critical Care, University College Hospital, and supported by researchers at the National Institute for Health Research University College London Hospitals Biomedical Research Centre.
ORCID iD: Dorothy Wade https://orcid.org/0000-0001-6431-3776
References
- 1.Coomber S, Todd C, Park G, et al. Stress in UK intensive care unit doctors. Br J Anaesth 2002; 89: 873–881. [DOI] [PubMed] [Google Scholar]
- 2.Embriaco N, Azoulay E, Barrau K, et al. High level of burnout in intensivists. Am J Respir Crit Care Med 2007; 175: 686–692. [DOI] [PubMed] [Google Scholar]
- 3.Poncet MC, Toullic P, Papazian L, et al. Burnout in critical care nursing staff. Am J Respir Crit Care Med 2007; 175: 698–704. [DOI] [PubMed] [Google Scholar]
- 4.Teixeira C, Ribeiro O, Fonseca AM, et al. Burnout in intensive care units – a consideration of the possible prevalence and frequency of new risk factors: a descriptive correlational multicentre study. BMC Anesthesiol 2013; 13: 38. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Medscape. Physician Lifestyle Report 2018, http://www.medscape.com/sites/public/lifestyle/2018 (2018, accessed 9 January 2019).
- 6.Laporta D, Burns J, Doig CJ. Bench-to-bedside review: dealing with increased intensive care staff turnover: a leadership challenge. Crit Care 2005; 9: 454–458. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Moss M, Good VS, Gozal D, et al. An Official Critical Care Societies Collaborative Statement: burnout syndrome in critical care health care professionals: a call for action. Am J Crit Care 2016; 25: 368–376. [DOI] [PubMed] [Google Scholar]
- 8.Intensive Care Society. Promoting wellbeing and reducing the risk of burnout in critical care. A joint ICS/FICM/BACCN working group, http://www.ics.ac.uk/ICS/Education/Wellbeing/ICS/Wellbeing.aspx?hkey=92348f51-a875-4d87-8ae4-245707878a5c (2018, accessed 8 January 2019).
- 9.Mealer M, Conrad D, Evans J, et al. Feasibility and acceptability of a resilience training program for intensive care nurses. AJCC 2014; 23: 97–105. [DOI] [PubMed] [Google Scholar]
- 10.Steinberg B, Klatt M, Duchemin AM. Feasibility of a mindfulness-based intervention for surgical intensive care unit personnel. Am J Crit Care 2017; 26: 10–18. [DOI] [PubMed] [Google Scholar]
- 11.West C, Dyrbye LN, Erwin PJ, et al. Interventions to prevent and reduce physician burnout: a systematic review and meta-analysis. Lancet 2016; 388: 2272–2281. [DOI] [PubMed] [Google Scholar]
- 12.Health and Safety Executive. Health and Safety Executive (HSE) management standards indicator tool,http://www.hse.gov.uk/stress/standards/pdfs/indicatortool.pdf (accessed 10 December 2012).
- 13.Ruotsalainen JH, Verbeek JH, Marine A, et al. Preventing occupational stress in healthcare workers. Cochrane Database Syst Rev 2015; 4: CD002892. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Dewe PJ, O’Driscoll MP, and Cooper CL. Factors influencing stress management interventions. In: Dewe PJ, O’Driscoll MP, and Cooper CL (eds) Coping with work stress: a review and critique. 1st ed. Oxford: Wiley-Blackwell, 2010, pp.134–144.
- 15.Van Schijndel RJM, Burchardi H. Bench-to-bedside review: leadership and conflict management in the intensive care unit. Crit Care 2007; 11: 234. [DOI] [PMC free article] [PubMed] [Google Scholar]