Healthy Work Environments and Nurse Burn Out (4)
The purpose of this paper is to provide a critique of an article on the topic of work environments and nurse burn out. The article to be critiqued is a large cross-sectional Swedish study that examines the associations among nurse practice environment, family/work conflict and burnout (Leineweber, et al., 2014). The intent of this critique is to provide evidence that the nurses’ working environment has an impact of personal psychosocial factors. The data for the study was retrieved from the survey-based Swedish section of the RN4CAST, which was multinational EU 7th Framework project. The original project involved RNs employed in surgical and medical inpatient health care. The paper will address the ethical protection of the participants, data collection method, data management and analysis, findings and implications for nursing practice, and conclusions.
1. Ethical Protection of Human Participants
Leineweber, et al., (2014) state that their study was reviewed and approved by the appropriate Stockholm Research Ethics Committee. As for the original RN4CAST study, Leineweber, et al., (2014) report that RNs were approached by a member of the Swedish Association of Health Professionals and provided with information on the study. Further information was sent by post through the Statistics Sweden administration. According to accepted Swedish ethical standards in research, responding to the request for participation and the return of the survey instrument constituted informed consent. There was no further information on the ethical handling of the collected data, or the risks or benefits to the participants. Presumably the data was combined and no personal names were associated with the information. The description of the ethical standards was described in the original study and a reference for that study was provided (Sermeus, Aiken, Van den Heede, Rafferty, Griffiths, et al., 2011).
2. Data Collection
The study variables were clearly defined and consisted of working environment, burnout, and family/work conflict. The types of information collected on the individual RN level were demographics, family and work conflicts, and duration of work experience. The family/work conflict was assessed by one question on the extent to which work negatively affected the RN’s personal life. The work environment was assessed on the department level using the Practice Environment Scale of Nursing Work Index, which is a validated and frequently-used instrument (Aiken, Sermeus, Van den Heede, Sloane, Busse, et al., 2012). Burnout was assessed using the Maslach Burnout Inventory Human Service Survey (Maslach, Jackson & Leiter, 1996). The Maslach survey was translated into Swedish and validated.
The data collection period for the survey was provided in the original study (Sermeus, Aiken, Van den Heede, Rafferty, Griffiths, et al., 2011). The sequence of data collection events for the participants was: (1) contact in the hospital by a member of the Swedish Association of Health Professionals and provided information on the study, (2) additional information and the survey were sent to the RNs by post in February 2010 and (3) completion and return of the survey. The rationale for the data collection method and additional information on the time line for data collection was available from the original study (Sermeus, Aiken, Van den Heede, Rafferty, Griffiths, et al., 2011).
3. Data Management and Analysis
The database consisted of self-reported survey information for 8, 620 RNs from 369 departments in 53 acute care Swedish hospitals. All RNs were directly involved in inpatient medical and surgical care. Data was segregated into three levels: individual nurse, department, and institution. The concept of burnout was operationalized as the three factors assessed in the scale: emotional exhaustion (9 items), personal accomplishment (8 items), and depersonalization (5 items). The statistical analyses used multilevel modelling in order to explain the variation in the outcome variable of burnout. The lack of specific information regarding department and organizational characteristics was made up by aggregating scores on appropriate items on the Practice Environment Scale of Nursing Work Index. The scale met the statistical criterial for aggregating data.
Three separate models were specified in the multilevel modelling process. Model 1 adjusted for the individual demographic data and the family/work conflict. Model 2 adjusted solely for the departmental variables aggregated from the Practice Environment Scale of Nursing Work Index. Model 3 adjusted for both the individual level and the departmental level variables. The odds ratios were calculated at a 96% confidence interval. Using the Ime4 statistical package, the models were compared in order to identify the best fit. The authors did not mention a specific paper trail of decisions; however, the statistical analyses were clearly described at each stage and all analyses were conducted with statistical software.
There was no need to provide measures to guard against researcher bias in the Leineweber, et al. (2014) study as all the data was in the database. However, in the original study (Sermeus, Aiken, Van den Heede, Rafferty, Griffiths, et al., 2011) a researcher bias was introduced in the question that assessed the family/work conflict. The question was worded to assess the extent to which work negatively affects the RN’s private life, with the available responses ranging from a great degree to a small degree. The phrasing assumes that the practice of nursing will necessarily have a negative influence. There was no field for responding that work had no impact on personal life.
4. Findings/Interpretation of Findings: Implications for Practice and Future Research
Leineweber, et al. (2014) found that the nurses had high levels of personal accomplishment and lower levels of depersonalization and emotional exhaustion. However, the one-third of the nurses could be classified as emotionally exhausted. There was a considerable between- department difference in the three burnout variables. Model 3 provided the better explanation for the associations of the individual, department and organization level variables, which suggested that organizational level variables were responsible for the variation in nurse burnout. The findings were presented in a logical manner and were easy to follow and understand. Although the relationship between burnout and family/work conflict is well-established in the nursing literature, it is difficult to come to any conclusions regarding the extent to which the study results can be generalized to all nursing practice reality because the participants and institutions were all European. Also, the demographic details did not cover critical factors such as single parenthood and number of children at home, which would impact family/work conflict.
The study limitations identified by the authors was lack of detail on the department level, which did not permit the inclusion of factors such as shift rotation and support from management. In addition, the cross-sectional design did not allow the authors to determine the direction of the impact, for example, does family/work conflict to burnout or vice versa.
The implications for nursing practice are on the departmental level and suggested that sufficient staffing, and management issues such as support and leadership are important for the mental health of nurses. The current reality for Europe is the same for other parts of the developed world, and that is the reducing levels of funding. Good leadership is paramount in a climate of economic restraints that place a burden on the ability of the department to provide adequate staffing. The authors suggested that further studies that provide greater detail on the individual level, such as family composition and shift work, and on the lower levels of management.
5. Conclusion
In conclusion, the study supports the relationship between nurse burnout and characteristics of the organization. The findings highlight the need for management to develop policies that facilitate the balance between work and family. Supervisors should apply strategies that reduce the impact of work on family responsibilities. The study has a large number of participants, which makes the statistics robust. The study demonstrated the relationship between emotional exhaustion and family/work conflicts, but not the other burnout factors of depersonalization and personal accomplishment. Therefore, the study does provide support for the importance of nurse-centered hospital policies. However, the lack of detail on the department and institution level and the necessity of aggregating the data weaken the utility of the findings.
References
Aiken, L. H., Sermeus, W., Van den Heede, K., Sloane, D.M., Busse, R., et al. (2012). Patient safety, satisfaction, and quality of hospital care: cross sectional surveys of nurses and patients in 12 countries in Europe and the United States. BMJ, 344: e1717.
Leineweber, C. Westerlund, H., Chungkham, H. S., Lindqvist, R., Runesdotter, S. & Tishelman, A. (2014). Nurses’ Practice Environment and Work-Family conflict in relation to burn out: A multilevel modelling approach. PLoS One, 9 (5), e96991. DOI:10:1371/journal.pone.0096991.
Maslach, C., Jackson, S. E., & Leiter, M. P. (1996). Maslach Burnout Inventory Manual. Palo Alto: Consulting Psychologists Press, Inc.
Sermeus, W., Aiken, L. H., Van den Heede, K., Rafferty, A. M., Griffiths, P., et al. (2011). Nurse forecasting in Europe (RN4CAST): Rationale, design and methodology. BMC Nursing, 10: 6.
