Safety Climate, Emotional Exhaustion and Job Satisfaction Among Brazilian Paediatric Professional Nurses
Safety Climate, Emotional Exhaustion and Job Satisfaction Among Brazilian Paediatric Professional Nurses
ALVES D.F.S. & GUIRARDELLO E.B. (2016) Safety climate, emotional exhaustion and job satisfaction
among Brazilian paediatric professional nurses. International Nursing Review 00, 000–000
Background: International studies indicate that job satisfaction and burnout interfere with the safety
climate and quality of care. However, no evidence of such relationships is available for Brazilian paediatric
hospitals.
Aims: To assess the correlation and predictive effect of emotional exhaustion and job satisfaction on the
perception of professional nurses at paediatric hospitals regarding safety climate and quality of care.
Design: Cross-sectional correlational design.
Methods: The study was conducted with registered nurses, technician and assistant nurses from two
Brazilian paediatric hospitals over 3 months in 2013–2014 using instruments to assess safety climate, quality
of care, job satisfaction and emotional exhaustion.
Results: Data related to 267 professional nurses from 15 inpatient wards and 3 intensive care units were
analysed. Overall, the respondents exhibited moderate emotional exhaustion, were satisfied with their jobs
and considered the quality of care as good. However, the respondents exhibited low concordance as to the
positive perception of the safety climate. The variables, emotional exhaustion and job satisfaction, exhibited
significant correlations with safety climate and were considered predictive of the latter.
Conclusion: Emotional exhaustion and job satisfaction among professional nurses influence the safety
climate at paediatric hospitals.
Implications for nursing or health policy: Investments to reduce emotional exhaustion and to improve
job satisfaction among professional nurses allocated to paediatric hospitals might contribute to the patients’
safety.
Keywords: Brazil, Burnout, Job Satisfaction, Nursing Staff, Patient Safety, Paediatric Hospitals, Professional,
Quality of Health Care
Correspondence address: Daniela Fernanda dos Santos Alves, 275 Julita de Souza Sampaio, Porto Feliz, S~ao Paulo 18540-000, Brazil; Tel: 55-19-992-609-029;
E-mail: [Link]@[Link].
Funding
Grant #2013/09441-0, S~ao Paulo Research Foundation (FAPESP).
Conflict of interest
No conflicts of interest have been declared by the authors.
Study aim Three hundred and twenty-four professional nurses were avail-
The aim of this study was to assess the correlation between able for the study, of which 267 returned the completed ques-
safety climate, quality of care, emotional exhaustion and job tionnaires, corresponding to a response rate of 82.4%.
satisfaction among professional nurses in paediatric care
units. The predictive effects of emotional exhaustion and job Measures
satisfaction on safety climate and quality of care were also The variable emotional exhaustion, the main component of
analysed. burnout syndrome, was analysed using Maslach Burnout
Inventory (MBI). The original version of MBI was composed
Design by Maslach & Jackson (1981) in order to assess physical and
This was a non-experimental study with cross-sectional corre- emotional exhaustion due to work. The Brazilian version of
lational design which used self-administered instruments for MBI, used in this study, was translated and adapted by
data collection. Tamayo (1997). Emotional exhaustion was measured using
the nine items of MBI where the respondents answered on a
Respondents 5-point Likert scale (never to always). Items include questions
The study was conducted with professional nurses working in such as ‘I feel emotionally drained from my work’ and ‘I feel
15 inpatient wards and 3 intensive care units of two paedi- fatigued when I get up in the morning and have to face
atric hospitals: one public (105 beds) and one private (136 another day on the job’. The scores of emotional exhaustion
beds), in the city of S~ao Paulo (Brazil). were classified based on tertiles (Aiken & Patrician 2000;
In Brazil, the professional nurses are divided into three cat- Tamayo 1997); for this study, values equal to or lower than
egories: registered nurses (RN), technicians and assistant 18.66 were considered as indicative of low levels of exhaus-
nurses. The RN has a bachelor degree and is responsible for tion; higher than 18.66 and equal to or lower than 24, as
highly complex tasks and care management. Technicians and moderate levels; and higher than 24, as high levels (Aiken &
assistant nurses have secondary school qualifications plus Patrician 2000; Tamayo 1997). In the current study, the
specific training of 1 year (assistant nurses) to 2 years (tech- Cronbach’s alpha for emotional exhaustion subscale was 0.87.
nician nurses). They are involved in low complexity care such The variables job satisfaction and safety climate were
as hygiene and comfort. Thus, assistants, technicians and reg- assessed using the Safety Attitudes Questionnaire – Short
istered nurses involved in direct patient care were considered form 2006 (SAQ). The SAQ, developed by Sexton et al.
to be eligible to participate in the study. (2006), was validated for Brazilian culture by Carvalho & Cas-
The sample size was calculated considering estimates of siani (2012). It contains 41 items and six subscales which
correlation coefficients among the study variables (emotional measure the professionals0 perception of patient safety issues.
exhaustion, job satisfaction, quality of care and safety climate) For this study, a safety climate subscale (seven items) and a
which were obtained in a previous study (Profit et al. 2014). job satisfaction subscale (five items) were used. The subscale
We considered the correlation coefficient 0.0 as null hypothe- safety climate evaluated the professionals’ perception about
sis, with significance level of 5% and sample power of 80%. the organizational commitment to patient safety and contains
The largest sample size obtained was of 215 respondents to items such as ‘I would feel perfectly safe being treated in this
estimate a correlation coefficient between emotional exhaus- unit’ and ‘I know the proper channels to direct questions
tion and quality of care different from 0.0. regarding patient safety in this unit’. The subscale job satisfac-
Data were collected from December 2013 to February 2014 tion was used to evaluate a positive working experience such
by the principal investigator and a student previously trained as ‘I like my job’ and ‘I am proud to work at this hospital’.
for this purpose. The official list of the professionals in all units The SAQ items are answered on 5-point Likert scales
and shifts from both institutions was obtained, including infor- (strongly disagree, slightly disagree, neutral, slightly agree and
mation on their length of work at the unit, personal leaves, strongly agree) and the respondents could also select option
vacations and sick leaves. The professional nurses who were on (X), not applicable. The subscale scores were calculated by
vacation or away for any reason were not considered when the average response of the items and it varied from 14.29 to
obtaining the sample. The potential respondents were informed 100.0 for safety climate and from 15.0 to 100.0 for job satis-
as to the voluntary nature of participation, the anonymity and faction. Scores over 75 indicated that the environment was
confidentiality of the data. The professionals who agreed to safe for patients and that the professional nurses were satis-
participate signed an informed consent form. All question- fied with their job. Any respondent who answered ‘not appli-
naires were collected within 2 weeks of having been delivered. cable’ to more than 20% of the questions was excluded from
the analysis (Carvalho & Cassiani 2012; Sexton et al. 2006). Table 1 Demographic characteristics of the study participants
In the present study, the Cronbach’s alpha was 0.62 for safety
climate and 0.86 for job satisfaction subscale. Characteristics (n = 267) n %
The quality of care was assessed using a questionnaire
derived from previous studies (Aiken et al. 2012; Gasparino Private hospital 136 50.9
et al. 2011). The respondents were asked to comment on the Public hospital 131 49.1
quality of the care provided at their units, with response Inpatient unit 187 70.0
Paediatric ICU 80 30.0
options on a 4-point Likert scale, varying from very poor to
Nurses 99 37.1
very good. In addition, the respondents were requested to Nursing assistants and technicians 168 62.9
provide information on the hospital type (public versus Female 245 91.8
private), unit type (inpatient versus intensive care), training, Married 144 53.9
gender, marital status, age, length of professional experience, Secondary school* 168 62.9
length of work at the unit and other types of employment. Higher education† 23 8.6
Non-degree graduate programmes‡ 76 28.5
No other type of employment 206 77.1
Ethical considerations
Good/very good quality of care 261 97.7
Introduction letters were sent by e-mail to the nursing man-
High level of emotional exhaustion 73 27.4
agement and research committees of both participating insti-
tutions. Both institutions gave authorization, and the study
*Secondary formation and additional specific training in nursing activities.
was approved by the research ethics committees of the hospi-
†Registered nurses with higher graduate level.
tals (approval numbers 347.759 and 353.171). The informed ‡Registered nurses with higher graduate level plus specialization.
consent was obtained from each participant of this study.
Data analysis unit was 4.5 years (SD: 5.3). The assistant and technician
The data were analysed using the software SAS 9.3 (Statistical nurses had an average of 4.2 patients under their care (SD:
Analysis System; SAS Institute Inc., Cary, NC, USA). Spear- 1.6), and the registered nurses had 13.4 patients (SD: 5.7)
man’s correlation coefficient was calculated to assess the cor- and 1–12 subordinates. Overall, the level of emotional
relation between the variables. Multiple linear regression was exhaustion of the sample was moderate; 27.3% exhibited a
used to test the predictive value of the variables emotional high burnout. The results corresponding to job satisfaction
exhaustion and job satisfaction relating to safety climate and were the opposite, as the average score was considered
perceived quality of care. For this analysis, age, gender, mari- favourable to safety. The average score relating to safety cli-
tal status, function (RN vs. assistant and technician nurses), mate was less than 75 and was thus indicative of low con-
training and experience in unit (number of years in unit) and cordance of the respondents with the questions assessing
in profession (number of years as a professional nurse) were patient safety (Table 2). Table 2 describes the correlation
considered as the control variables for this study. The step- between the variables emotional exhaustion, job satisfaction
wise method was used for variable selection. The level of sig- and safety climate. All correlations were significant and indi-
nificance was set to 5% in all the analyses. cate that the lower the level of emotional exhaustion, the
more positive the perception of the safety climate and the
Results higher the job satisfaction. In addition, they indicate that
A total of 267 professional nurses participated in this study, the more satisfied the respondents were with their job, the
with an average age of 34.9 years old (standard deviation – more positive the safety climate.
SD: 7.9). Most were female, worked as assistant and techni- The tests confirmed that only the job satisfaction variable
cian nurses, had an average workload of 43 h per week was predictive of safety climate. The regression coefficients
(SD: 14.1), had no other type of employment and have at indicate that a 1-point unit increase in the job satisfaction
least secondary school level qualifications. The quality of score may increase the safety climate scores by 45%, which
care was reported as being predominantly good or very could be applied to 27% of the sample (Table 3). The vari-
good. Table 1 presents the descriptive analysis of the inves- able quality of care could not be tested either as to its correla-
tigated variables. tion or in the regression model due to the lack of variability
The average length of professional experience was in the responses because the number of ‘poor’ or ‘very poor’
8.8 years (SD: 6.5), and the average length of work at the responses was too small.
Table 2 Means, minima, maxima, standard deviations, reliabilities and correlations of the study variables
*P < 0.0001.
In addition, the results of this study have demonstrated and has not been used because it could create bias in the
that increasing the professionals’ job satisfaction might result analyses conducted.
in a work climate favourable for patient safety. Changes in The results evidence the strong influences of organizational
work environment, such as providing the opportunity to dis- and work environment factors on patient safety. Studies
cuss issues about the patient care (Van Bogaert et al. 2014a), including the variables analysed in this study and the occur-
are strategies that aim to improve the professionals0 job satis- rence of adverse events might be the starting points for signif-
faction, and might favour patient safety (Profit et al. 2014; icant changes in patient safety.
Rochefort & Clarke 2010).
Regarding the quality of care, most respondents rated this Implications for nursing and health policy
as good or very good. However, one essential factor should The principal motivation for this study was a perception of
be considered when interpreting this result. The quality of the efforts of nurse managers in improving the patients’ safety
care was evaluated by only one question, which was self- and the quality of care in Brazilian healthcare institutions.
reported and therefore might have led to a response bias. The results challenge the organizations to implement changes
in the work environment in order to reduce emotional
Limitations exhaustion and improve the professionals’ satisfaction with
This study was conducted at two paediatric hospitals accred- their job.
ited by official agencies, thus, the external validity of the Interventions to improve work environment and to reduce
results is limited, as they do not correspond to the reality of burnout rate, in addition to improvements of the work pro-
most hospitals caring for children and adolescents in Brazil. cess and the clinical management of work, might contribute
The factors associated with the large number of respon- to reduce the occurrence of adverse events (Van Bogaert et al.
dents who reported high levels of emotional exhaustion were 2014a,b,c). In the units where opportunities for dialogue are
not investigated. It is worth observing that the pattern of dis- frequent and the bedside rounds include nurses and nurse
tribution of the responses related to the variable quality of managers, the work environment becomes more favourable to
care limited the correlation and regression analyses. Future professional practice, the burnout rate of nursing staff
studies on the work environment using instruments to assess decreases and there is an improvement of patient safety (Sex-
the quality of care might be useful to test a causal model for ton et al. 2014).
such relationships.
Author contributions
Recommendations for future studies Study design: DFSA, EBG.
In future studies, it is recommended investigating the corre- Data collection: DFSA.
lation between the number of hours worked, the occurrence Data analysis: DFSA, EBG.
of emotional exhaustion and the implications of these on Study supervision: EBG.
patient safety. In addition, the fact that the analysed profes- Manuscript writing: DFSA.
sionals were relatively young and had less than 10 years of Critical revisions for important intellectual content: EBG.
professional experience must also be taken into considera-
tion. It is worth observing that the present study used a
References
mixed sample, including professionals with various levels of
Aiken, L.H. & Patrician, P.A. (2000) Measuring organizational traits of
training, and these results might not have been the same
hospitals: the Revised Nursing Work Index. Nursing Research, 49 (3),
had the various professional categories been analysed sepa- 146–153. doi:10.1097/00006199-200005000-00006.
rately. Aiken, L.H., et al. (2012) Patient safety, satisfaction, and quality of hospi-
tal care: cross sectional surveys of nurses and patients in 12 countries
Conclusion in Europe and the United States. BMJ, 344, e1717. doi:10.1136/
In the hospitals studied, the results suggest that investments bmj.e1717.
that aim to reduce the emotional exhaustion and improve the Aranaz-Andres, J.M., et al. (2011) Prevalence of adverse events in the
job satisfaction of professional nurses might contribute hospitals of five Latin American countries: results of the ‘Iberoamerican
favourably and substantially to improvements in the safety study of adverse events’ (IBEAS). BMJ Quality and Safety, 20 (12),
climate. This study was not able to assess the effects of those 1043–1051. doi:10.1136/bmjqs.2011.051284.
variables on the quality of care as perceived by the nursing Ausserhofer, D., et al. (2013) The association of patient safety climate
staff. The variable quality of care did not show data variability and nurse-related organizational factors with selected patient outcomes:
a cross-sectional survey. International Journal of Nursing Studies, 50 (2), Kirwan, M., Matthews, A. & Scott, P.A. (2013) The impact of the work
240–252. doi:10.1016/[Link].2012.04.007. environment of nurses on patient safety outcomes: a multi-level
Bates, D.W., Larizgoitia, I., Prasopa-Plaizier, N. & Jha, A.K. (2009) Global modelling approach. International Journal of Nursing Studies, 50 (2),
priorities for patient safety research. BMJ, 338, b1775. doi:10.1136/ 253–263. doi:10.1016/[Link].2012.08.020.
bmj.b1775. Landrigan, C.P. (2011) Healthcare provider working conditions and well-
Beal, A.C., et al. (2004) Quality measures for children’s health care. being: sharing international lessons to improve patient safety. Jornal de
Pediatrics, 113 (Suppl. 1), 199–209. Pediatria, 87 (6), 463–465. doi:10.2223/JPED.2147.
Carvalho, R.E.F. & Cassiani, S.H.B. (2012) Cross-cultural adaptation of Maslach, C.M. & Jackson, S.E. (1981) The measurement of experienced
the Safety Attitudes Questionnaire – Short Form 2006 for Brazil. burnout. Journal of Occupational Behaviour, 2 (2), 99–113.
Revista Latino-Americana de Enfermagem, 20 (3), 575–582. doi:10.1590/ Panunto, M.R. & Guirardello, E.B. (2013) Professional nursing practice:
S0104-11692012000300020. environment and emotional exhaustion among intensive care nurses.
Cimiotti, J.P., Aiken, L.H., Sloane, D.M. & Wu, E.S. (2012) Nurse staff- Revista Latino-Americana de Enfermagem, 21 (3), 765–772. doi:10.1590/
ing, burnout, and health care-associated infection. American Journal of S0104-11692013000300016.
Infection Control, 40 (6), 486–490. doi:10.1016/[Link].2012.02.029. Profit, J., et al. (2014) Burnout in the NICU setting and its relation to
Dawson, A.J., et al. (2014) Nursing churn and turnover in Australian safety culture. BMJ Quality and Safety, 23, 806–813, 1–8 doi:10.1136/
hospitals: nurses perceptions and suggestions for supportive strategies. bmjqs-2014-002831.
BMC Nursing, 13 (1), 11. doi:10.1186/1472-6955-13-11. Riehle, A., Braun, B.I. & Hafiz, H. (2013) Improving patient and worker
Duffield, C., et al. (2007) Glueing It Together: Nurses, Their Work Envi- safety: exploring opportunities for synergy. Journal of Nursing Care
ronment and Patient Safety. Centre for Health Services Management, Quality, 28 (2), 99–102. doi:10.1097/NCQ.0b013e3182849f4a.
Sydney, NSW. Rochefort, C.M. & Clarke, S.P. (2010) Nurses’ work environments, care
Duffield, C., et al. (2011) Nursing staffing, nursing workload, the work rationing, job outcomes, and quality of care on neonatal units. Journal
environment and patient outcomes. Applied Nursing Research, 24 (4), of Advanced Nursing, 66 (10), 2213–2224. doi:10.1111/j.1365-
244–255. doi:10.1016/[Link].2009.12.004. 2648.2010.05376.x.
Ebright, P. (2014) Patient safety in the current health care environment: Sexton, J.B., et al. (2006) The safety attitudes questionnaire: psychometric
complexity of work remains an essential component. Western Journal of properties, benchmarking data, and emerging research. BMC Health
Nursing Research, 36 (7), 851–854. doi:10.1177/0193945914540576. Services Research, 6, 44. doi:10.1186/1472-6963-6-44.
Feng, X.Q., et al. (2011) The relationship between management safety Sexton, J.B., et al. (2014) Exposure to Leadership WalkRounds in neona-
commitment and patient safety culture. International Nursing Review, tal intensive care units is associated with a better patient safety culture
58, 249–254. doi:10.1111/j.1466-7657.2011.00891.x. and less caregiver burnout. BMJ Quality and Safety, 23 (10), 814–822.
Gasparino, R.C., Guirardello, E.B. & Aiken, L.H. (2011) Validation of the doi:10.1136/bmjqs-2013-002042.
Brazilian version of the Nursing Work Index-Revised (B-NWI-R). Singer, S.J. & Vogus, T.J. (2013) Reducing hospital errors: interventions
Journal of Clinical Nursing, 20 (23–24), 3494–3501. doi:10.1111/j.1365- that build safety culture. Annual Review of Public Health, 34, 373–396.
2702.2011.03776.x. doi:10.1146/annurev-publhealth-031912-114439.
Groves, P.S., Meisenbach, R.J. & Scott-Cawiezell, J. (2011) Keeping Speroff, T., et al. (2010) Organisational culture: variation across hospitals
patients safe in healthcare organizations: a structuration theory of and connection to patient safety climate. Quality and Safety in Health
safety culture. Journal of Advanced Nursing, 67 (8), 1846–1855. Care, 19 (6), 592–596. doi:10.1136/qshc.2009.039511.
doi:10.1111/j.1365-2648.2011.05619.x. Tamayo, M.R. (1997) Relationship Between the Burnout Syndrome
Jacobs, L.M., Nawaz, M.K., Hood, J.L. & Bae, S. (2012) Burnout among and Organizational Values in the Nursing Staff of Two Public
workers in a pediatric health care system. Workplace Health and Safety, Hospitals. Master’s Thesis. Universidade de Brasılia, Brazil.
60 (8), 335–344. doi:10.3928/21650799-20120726-03. [in portuguese].
Jha, A.K., Prasopa-Plaizier, N., Larizgoitia, I. & Bates, D.W. (2010) Teng, C.I., et al. (2010) Interactive effects of nurse-experienced time
Patient safety research: an overview of the global evidence. Quality and pressure and burnout on patient safety: a cross-sectional survey.
Safety in Health Care, 19 (1), 42–47. doi:10.1136/qshc.2008.029165. International Journal of Nursing Studies, 47 (11), 1442–1450.
Kaushal, R., Barker, K.N. & Bates, D.W. (2001) How can information doi:10.1016/[Link].2010.04.005.
technology improve patient safety and reduce medication errors in Van Bogaert, P., et al. (2014a) Impact of role-, job- and organizational
children’s health care? Archives of Pediatrics and Adolescent Medicine, characteristics on Nursing Unit Managers. Journal of Advanced Nursing,
155 (9), 1002–1007. doi:10.1001/archpedi.155.9.1002. 70 (11), 2622–2633. doi:10.1111/jan.12449.
Kelly, L.A., McHugh, M.D. & Aiken, L.H. (2011) Nurse outcomes in Van Bogaert, P.V., Dilles, T., Wouters, K. & Van Rompaey, B.V. (2014b)
Magnetâ and non-magnet hospitals. Journal of Nursing Administration, Practice environment, work characteristics and levels of burnout as
41 (10), 428–433. doi:10.1097/NNA.0b013e31822eddbc. predictors of nurse reported job outcomes, quality of care and patient
adverse events: a study across residential aged care services. Open Journal of Nursing Studies, 51 (8), 1123–1134. doi:10.1016/[Link].
Journal of Nursing, 4 (5), 343–355. doi:10.4236/ojn.2014.45040. 2013.12.009.
Van Bogaert, P., et al. (2014c) Nursing unit teams matter: impact of Wilson, B.L. (2010) Keeping an eye on patient safety using human factors
unit-level nurse practice environment, nurse work characteristics, engineering (HFE): a family affair for the hospitalized child. Journal for
and burnout on nurse reported job outcomes, and quality of care, Specialists in Pediatric Nursing, 15 (1), 84–87. doi:10.1111/j.1744-
and patient adverse events – a cross-sectional survey. International 6155.2009.00221.x.