Burnout is a psychological, work-related syndrome; a result of long-term exposure to emotional and interpersonal stressors in the workplace, manifested as negative self-esteem, a negative attitude towards work, and diminished interest in the clients/patients. Within the syndrome, three dimensions have been identified: emotional exhaustion (EE), depersonalization (D), and personal accomplishment (PA). Nurses working in high-stress areas, such as critical care, report high levels of burnout. Hospital nurses have higher burnout scores than those working in other settings, due to the emotional and organisational demands creating the perception of a excessive workload, and stressors associated with characteristics related to the work environment.
The aims of my research were to explore the associations between burnout syndrome, coping mechanisms and job satisfaction in critical care nurses, and to explore attitudes towards and a sense of knowledge about burnout syndrome in critical care nurses who had been identified as burnt out (as shown by their total MBI score). The following goals were reached: (i) to determine the incidence of burnout syndrome in critical care nurses; (ii) to explore the associations between gender and burnout, and between burnout syndrome, job satisfaction and coping mechanisms in critical care nurses; (iii) to present the associations between burnout syndrome and all these characteristics in a multivariate modelling process; and (iv) to explore attitudes towards and knowledge of the factors known to be related to burnout in burnt out critical care nurses.
A quantitative cross-sectional multi-centre study was conducted and a phenomenological theoretical framework was used in the qualitative part of the study, with semi-structured interviews.
Participants and procedure
The nurses were asked to voluntarily participate in the study. In the quantitative part, a convenience sampling method was used, and the target population was critical care nurses employed in the Intensive Care Units (ICUs) of five Croatian university hospitals. Data collection was carried out from April to September 2017, and 620 participants were recruited (544 female (87.7%) and 76 male (12.3%)). The inclusion criterion for this study was more than six months of work experience.
During the year that the data were collected, the approximate number of critical care nurses in Croatia was 3500, of which approximately 13.0% were male nurses. It is therefore safe to conclude that the gender structure of this study sample reflects the gender structure of the population of critical care nurses in Croatia.
The qualitative part of the study had an exploratory phenomenological design, and was carried out in several Croatian ICUs in 2017. A sample of ICU nurses with burnout, according to their score on the Maslach Burnout Inventory, were chosen randomly from five hospitals and participated voluntarily. Of the 28 participants, 86% were women (n=24) and 14% men (n=4), most aged 36-45 (n=11 (40%)) and 26-35 years (n=10 (36 %)).
Several demographic characteristics were collected, i.e. age, gender, education, duration of work in the ICU, and type of ICU. Aside from these, in the quantitative part of the study, the Maslach Burnout Inventory (MBI), the Ways of Coping Scale (WOC) and the Job Satisfaction Scale (JSS) were administered; the MBI showed good internal consistency reliability with the Cronbach α, ranging from 0.74 to 0.90 for each subscale; all the WOC scales showed moderate reliability of internal consistency (Cronbach α between 0.6 and 0.7); and for the JSS the Cronbach α was<0.80.
In the qualitative part of my research, semi-structured interviews were conducted up to the saturation point, and the conversations were audio-recorded and transcribed verbatim.
The data analyses were carried out using IBM SPSS Statistics for Windows (version 22.0). Descriptive statistics (frequency, percentage, mean, standard deviation) were used to summarise the main characteristics of the sample. The differences in the MBI, WOC and JSS were assessed using the independent t-test, the Mann-Whitney U test, and the Kruskal-Wallis test. Data normality was tested by the Kolmogorov-Smirnov test.
For the WOC, the Kaiser-Meyer-Olkin measure of sampling adequacy gave a high value (KMO=0.876), and the Bartlett's test of sphericity was statistically significant (p<0.001), indicating the suitability of a factor analysis; by the rotated solution, F1 explained 25.9% of the initial variance, and F2 explained 24.5% of the initial variance; both factors cumulatively explained 50.4% of the initial variance. The internal consistency reliability for active coping (F1) was 0.80 and for passive coping (F2) was 0.75.
The associations between gender and burnout, coping mechanism and job satisfaction were examined using the Chi-square test. Gender, age and the coping and job satisfaction scales’ total scores were independent variables in the logistic regression modelling, with burnout (each dimension of MBI) being the dependent variable. Significance was set at p<0.05.
In the qualitative analysis, the initial codes were generated in the first stage of open coding, and the data were dealt with collaboratively and corroboratively; afterwards themes based on the codes were constructed, by organising the data into meaningful groups. A review of the initial codes and (re)combining them into previous/new themes took place before developing and defining the names of the themes. Each theme captured something important about the data in relation to the research question, and represented some level of patterned response or meaning within the data set, with the process corresponding to axial coding procedure. Finally, the number of themes was reduced to a more manageable set of main themes and followed by interpretation at the final stage.
In the quantitative part of the study, the majority of the sample were female nursing staff (87.7%), aged 26-35 (38.9%) with less than 5 years of work experience (39.4%); the primary work department was general surgical ICU (36.3%), and the major location was UHC Zagreb (51.8%). The participants were aged 33.5±7.7 and their work experience ranged from 1 to 389.5 years.
Of all the participants, 137 (22.1%) had a high score on EE; 49 (7.9%) had a high score on DP; 214 (34.5%) had a low score on PA; and 72 (11.6%) had burnout according to the total score of burnout (MBItot). A lower proportion of people with 5-10 years of work experience reported a high DP (15.5%), while the proportions with low PA and medium DP were 20.6% and 17.6%, respectively (p=0.003). Gender was not related to job satisfaction (p˂0.443), or to coping mechanisms (active coping p˂0.927 and passive coping p˂0.144). Active coping used somewhat was identified in 340 (62.5%) women and 48 (63.2%) men.
Where the participants reported high EE, passive coping was reportedly used quite a bit by a higher proportion (28.5%); otherwise 14.0% for low and 16.2% for medium EE (p=0.005) was found. More men reported medium (17.2%) or high (16.3%) DP, while more women reported low DP (p=0.045). Where the participants reported high DP, their JSS was likely to be satisfied (38.8%), but satisfied for JSS was reported even more often for those with low (63.8%) and medium (64.3%) DP (p<0.001). Where the participants reported high DP, passive coping was reportedly used quite a bit by a greater proportion (34.7%); otherwise 12.9% for low and 26.5% for medium DP (p<0.001) was found. Participants who scored lower on DP reported a higher JSS (p<0.001). A satisfied JSS (OR=4.04, 95% CI=1.81-9.03, p=0.001) and a very satisfied JSS (OR=10.40, 95% CI=1.81-9.03, p<0.001) were associated with a higher PA. Five to ten years of work experience (OR=0.56, 95% CI=0.33-0.95, p=0.032) were associated with a lower PA. Men reported a higher percentage of medium and high DP in comparison to women (p=0.045). Gender did not make a difference to burnout (MBITOT), EI or PA levels.
MBItot was used as a dependent variable in the logistic regression. Lower scores of MBItot were associated with job satisfaction assessed as neutral (OR=0.22, 95% CI=0.10-0.52, p=0.001), satisfied (OR=0.04, 95% CI=0.02-0.09, p<0.001) and very satisfied (OR=0.01, 95% CI=0.00-0.02, p<0.001). Higher scores on MBItot were associated with passive coping described as used somewhat (OR=3.08, 95% CI=1.39-6.83, p=0.006), used quite a bit (OR=9.93, 95% CI=4.01-24.61, p<0.001) and five to ten years of work experience (OR=1.99, 95% CI=1.10-3.60, p=0.024). Using MBItot as a measure of burnout in the modelling process was shown to be useful, and nearly 36% of the variance was explained (Nagelkerke R2=0.359).
With regard to burnout syndrome incidence in critical care nurses, the results of this study are comparable to the published research findings of other professions, so H1 was confirmed. The incidence in this study was EE 22.1%, D 7.9%, PA 34.5%, and TS 11.6%, while the literature shows it to vary between soldiers, EE 16.3%, D 8.64%, PA 30.7%, and physicians, EE 42.4%, D 16.0%, PA 15.2%. The H2 that burnout and job satisfaction are not gender-related was also confirmed, as was H3 that coping mechanisms and job satisfaction are associated with burnout syndrome.
In the qualitative analysis, several emergent themes, compromised private life, stressful work demands, stress reduction options, protective workplace measures and a sense of knowledge reflected a variety of experience, attitudes and knowledge of burnout. Nurses with burnout provided an insight into their experience and attitudes, and the problems burnout creates.
The results of the quantitative part of my research concerning burnout and the total score on the MBI are comparable with the results of studies of other professions, but they vary with regard to the sample and the work conditions of the countries involved. A work environment with preventive programmes showed less incidence of burnout.
The qualitative analysis showed an insufficient sense of knowledge about burnout in Croatian critical care nurses, and a need to implement education on burnout in the curriculums of nursing schools. Given that active coping may decrease the incidence of burnout, training and capacity building should acknowledge this finding. This study brought new knowledge and gave new direction for curriculum programmes in nursing education, which should include knowledge of and skills in burnout and coping strategies, i.e. information, awareness-raising, and specific guidelines on coping, burnout detection and prevention. Approaching burnout prevention through attitudes/social learning may be a novel and feasible model for addressing this issue.
Since level of education was found to be associated with a lower incidence of personal accomplishment, nurse management should encourage nurses to take part in further/continuous education.
Strengthening the job-satisfaction-related factors may reduce the impact of burnout in critical care nurses. With regard to the association between job satisfaction and the incidence of burnout, hospital management should consider improvements in the work environment. Preventive strategies should not be only for critical care nurses, but for all nurses in healthcare institutions, as well as other healthcare professionals, e.g. physicians, physiotherapists, and occupational therapists.
The professional benefit of this study includes the potential to introduce and adopt functional coping strategies. Given that active coping may decrease the incidence of burnout, training and capacity building should acknowledge this finding.
The information provided can help ICU nurses and nurses in general to understand how burnout affects their professional and family lives, and offers a variety of methods aimed at mitigating these harmful effects. Finally, the findings can motivate healthcare professionals with limited skills and/or knowledge of burnout to at least accept the necessity for modifying their coping strategies to be more problem-oriented and constructive.