Predictors of burnout Discussion.

Background: High levels of work-related stress, burnout, job dissatisfaction, and poor health are common within the nursing profession. A comprehensive understanding of nurses’ psychosocial work environment is necessary to respond to complex patients’ needs. The aims of this study were threefold: (1) To retest and confirm two structural equation models exploring associations between practice environment and work characteristics as predictors of burnout (model 1) and engagement (model 2) as well as nurse-reported job outcome and quality of care; (2) To study staff nurses’ and nurse managers’ perceptions and experiences of staff nurses’ workload; (3) To explain and interpret the two models by using the qualitative study findings.

ORDER CUSTOM, PLAGIARISM-FREE PAPER

Method: This mixed method study is based on an explanatory sequential study design. We first performed a cross- sectional survey design in two large acute care university hospitals. Secondly, we conducted individual semi-structured interviews with staff nurses and nurse managers assigned to medical or surgical units in one of the study hospitals. Study data was collected between September 2014 and June 2015. Finally, qualitative study results assisted in explaining and interpreting the findings of the two models.

Results: The two models with burnout and engagement as mediating outcome variables fitted sufficiently to the data. Nurse-reported job outcomes and quality of care explained variances between 52 and 62%. Nurse management at the unit level and workload had a direct impact on outcome variables with explained variances between 23 and 36% and between 12 and 17%, respectively. Personal accomplishment and depersonalization had an explained variance on job outcomes of 23% and vigor of 20%. Burnout and engagement had a less relevant direct impact on quality of care (≤5%). The qualitative study revealed various themes such as organisation of daily practice and work conditions; interdisciplinary collaboration, communication and teamwork; staff nurse personal characteristics and competencies; patient centeredness, quality and patient safety. Respondents’ statements corresponded closely to the models’ associations.

Conclusion: A deep understanding of various associations and impacts on studied outcome variables such as risk factors and protective factors was gained through the retested models and the interviews with the study participants. Besides the softer work characteristics — such as decision latitude, social capital and team cohesion — more insight and knowledge of the hard work characteristic workload is essential.

Keywords: Burnout, Work engagement, Job satisfaction, Turnover intentions, Quality of care, Structural equation model, Sensitizing concepts

* Correspondence: peter.vanbogaert@uantwerpen.be; peter.van.bogaert@uza.be 1Nursing and Midwifery Sciences, Centre for Research and Innovation in Care (CRIC), Faculty of Medicine and Health Sciences, University of Antwerp, Universiteitsplein 1, B-2610 Wilrijk, Belgium 2Department of Nursing, Antwerp University Hospital, Wilrijkstraat 10, B- 2650 Edegem, Belgium Full list of author information is available at the end of the article

© The Author(s). 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.

Van Bogaert et al. BMC Nursing (2017) 16:5 DOI 10.1186/s12912-016-0200-4

Background Thirty years of research on burnout and on nurse work en- vironment provide a body of knowledge about occupational stress and well-being and insight in the psychosocial work environment of nurses, one of the largest workforce in healthcare. Both research domains started empirically with a lack of theoretical frameworks. Research on burnout and psychosocial work environment has predominantly been conducted using the Maslach Burnout Inventory-Human Service Survey [33]. The primary themes in burnout research fit readily into the six areas of worklife such as workload, control, community, fairness, reward and value congruence [31]. Engagement as the positive pole of a con- tinuum and the opposite of burnout became an additional and interesting research domain to feature the person-job fit [41]. In line with Maslachs’ primary themes, Karasek and Theorell [25] have developed the job demand-control- support model that consists of three main dimensions: job demands, job decision latitude and job social support. This model provides insights about the mechanism of job related characteristics within specific nurse work environments such as emergency nursing, oncology nursing, mental health nursing and nurse unit managers [1, 2, 11, 16, 51]. Research on nurse work environment started with the observation that some hospitals in the US were more successful in attracting and retaining nurses compared to other hospitals. In addition, these researchers have been focused on to what extent certain relevant aspects were generalizable and transferable to other hospitals [34]. A substantial number of studies identified and linked aspects of a balanced, healthy and supportive psychosocial work environment ([20, 27, 30, 32]) with quality and patient safety indicators such as patient satisfaction, mortality, co-morbidity and adverse events [5, 6, 18]. Furthermore, intervention studies were conducted to evaluate quality improvement projects aiming practice environments that support highly motivated and skilled nurses answering accurately complex patient needs. In the US imple- mentations of ANCC Magnet Hospital key compo- nents including transformational leadership, structural empowerment, exemplary professional practice and new knowledge, innovations and improvements [7, 56]. In the UK and other European countries implementations of or the Productive Ward – Releasing Time to Care™ program [35, 49, 54, 55]. Our research program was initiated more than 10 years

ago, adapting these research insights and knowledge in the Belgian context and meanwhile aiming better under- standing of the associations between nurse practice environment and nurse work characteristics such as workload, decision latitude and social capital and outcome variables such as feelings of burnout and engagement, nurse reported job outcomes and quality of care [42, 43, 45, 46, 50]. Our research initiatives have been contributing

to a clear understanding of nurses their practice environ- ment that could support and guide the practice commu- nity. Therefore, this study based on an explanatory sequential design, was a next step in a series of studies that developed comprehensive models providing a deep understanding of various associations and impacts on studied outcome variables. The study aims were threefold: (1) To retest and confirm two structural equation models exploring associations between practice environment and work characteristics as predictors of burnout (model 1) and engagement (model 2) as well as nurse-reported job outcome and quality of care; (2) To study staff nurses’ and nurse managers’ perceptions and experiences of staff nurses’ workload; (3) To explain and interpret the two models by using the qualitative study findings.

Methods This mixed method study was based on an explanatory sequential study design [15]. The study started in a first phase with a quantitative approach collecting and analys- ing of quantitative data with the aim to retest and confirm two previous developed models. The second phase, a qualitative study, existed of collecting and analysing quali- tative data based on semi-structured interviews. Both study phases were conducted independently. Finally, in a third phase qualitative study results assisted in explaining and interpreting the findings of the two model.

Study population Quantitative data set The study was conducted in two acute care university hospitals, one in the Dutch- and one in the French- speaking part of Belgium, with 600 and 850 beds respectively. All participants were staff nurses working in direct care in either medical, surgical, obstetric, geri- atric or intensive care units and operating theatres including adult and paediatric care units. Participants were invited by one of the investigators to participate in the study on a voluntary basis. Data collection took place between September 2014 and May 2015. Respondents could complete the self-report questionnaires electronically either at home and/or in the hospital.

Qualitative data set The purpose of the qualitative study was to investigate staff nurses’ and nurse managers’ perceptions and expe- riences of staff nurses’ workload. To understand the complexity of staff nurses workload we included for this study a purposive sample with typical cases of staff nurses as well as nurse managers practicing on medical or surgical units. Assuming that medical and surgical nursing units are relatively comparable in terms of staff nurse practice environment and nurse work characteris- tics such as workload, we might expect similar perceptions

Van Bogaert et al. BMC Nursing (2017) 16:5 Page 2 of 14

and experiences. Each staff nurse and nurse manager of the participating units were invited by two study investiga- tors, respectively. Data were collected until sufficiency was obtained on the research topics (staff nurses = 9; nurse managers = 10). The semi-structured interviews were organized only in the Dutch-speaking university hospital between January 2015 and March 2015 and performed in a dedicated room. The hospital had recently implemented the Productive Ward programme and became involved in an accreditation process (JCI – Joint Commission Inter- national) as a part of a larger national hospital account- ability process.

Ethical considerations The institutional review board of each study hospital approved the qualitative study. In addition, a qualified ethics review committee (Antwerp University Hospital – University of Antwerp Belgium) approved the qualita- tive study.

Procedure and data analyses Quantitative study: model retesting and confirmation The two models were carefully developed and fitted sufficiently to a cross-sectional dataset based on survey de- sign. Moreover, we used a set of measurement instruments

Fig. 1 Model 1 – burnout as mediating outcome variable – retested model. Legend: All variables, with the exception of workload, emotional exhaustion and depersonalization were coded for analysis whereby higher scores indicated a stronger agreement or more favourable ratings. On the latter measures, higher scores are suggestive of unfavourable perceptions or conditions. All pathways were significant (p < .05). The independent variables of nurse practice environment predict the mediating variables of burnout dimensions, as well as job outcomes and nurse-assessed quality of care (dependent variables). In addition, workload, decision latitude, and social capital have a mediating position between the nurse practice environment and burnout dimensions. Nurse–physician relations and hospital management – organizational support impact nurse management at the unit level. Nurse management at the unit level has a strong direct impact on job outcomes and nurse-assessed quality of care as well as on decision latitude and social capital. Hospital management – organizational support has a direct impact on personal accomplishment and an indirect impact on the outcome variables through workload and burnout dimensions. Nurse–physician relations shows an indirect impact on the outcome variables through decision latitude. Social capital has an inverse impact on feelings of emotional exhaustion, and decision latitude supports feelings of personal accomplishments. Personal accomplishment, impacts indirectly by emotional exhaustion and directly by depersonalization, has a direct impact on job outcomes and nurse-assessed quality of care. The variances in job outcomes and nurse-assessed quality of care explained by this model were 63 and 53%, respectively. Nurse management at the unit level has a strong direct impact on outcome variables with explained variances of 25 and 36%, respectively

Predictors of burnout Discussion.