The Relationship between Social Support and Burnout among ICU Nurses in Shanghai:A Cross-Sectional Study

Li Li，Hong Run*，Wei-Jun Yun

aDental Department，The Ninth People's Hospital affiliated to Shanghai Jiaotong University School of Medicine，Shanghai 200011，China

bNursing Department，The Ninth People's Hospital affiliated to Shanghai Jiaotong University School of Medicine，Shanghai 200011，China

Original article

The Relationship between Social Support and Burnout among ICU Nurses in Shanghai:A Cross-Sectional Study

Li Lia，Hong Ruana*，Wei-Jun Yuanb

aDental Department，The Ninth People's Hospital affiliated to Shanghai Jiaotong University School of Medicine，Shanghai 200011，China

bNursing Department，The Ninth People's Hospital affiliated to Shanghai Jiaotong University School of Medicine，Shanghai 200011，China

A R T I C L E I N F O

Article history：

6 March 2015

Accepted 25 April 2015

Published 20 June 2015

Social support

Burnout

ICU nurses

Relationship

Objective:The associations between social support and burnout were explored in ICU nurses of Shanghai.

Methods:We performed a cross-sectional study of 356 ICU nurses by applying random cluster sampling.Data were collected using self-reported questionnaires under the instruction of trained investigators.Data on emotional exhaustion，depersonalisation and feelings of low personal accomplishment etc.were collected，calculated and analyzed.

Results:The participants had a mean age of 26.96 years（SD 4.07）.The mean value（M）and standard deviation（SD）of emotional exhaustion was M=31.85，SD=8.38，those of depersonalisation was

M=11.69，SD=5.54 and those of feelings of low personal accomplishment was M=19.79，SD=7.02. The high degree of emotional exhaustion（EE），depersonalisation（DP），and lack of personal accomplishment（PA）were revealed to be 76.4%，39.6%，and 94.9%，respectively.The major influencing factors of emotional exhaustion included support from co-workers（b=0.343，t=1.98，P=0.049），taking leave（b=-1.182，t=-3.747，P=0.001），requisition of work（b=-1.41，t=-2.369，P=0.018），and supervisor support（b=-0.524，t=-3.926，P=0.001）.The major influencing factors of depersonalisation were support from the supervisor（b=-0.333，t=-4.146，P=0.001），age（b=-0.89，t=-2.272，P= 0.024）and requisition of work（b=-0.148，t=-2.124，P=0.034）.There was a positive co-relationship between personal accomplishment and supervisor support.

Conclusions:Supervisor support，age，and requisition of work were the major influencing factors of depersonalisation.In addition，supervisor support plays an important role in low personal accomplishment.Further research should focus on supervisor support，co-worker support，time on leave，and requisition of work associated with emotional exhaustion.

ⓒ2015 Shanxi Medical Periodical Press.Production and hosting by Elsevier B.V.This is an open access article under the CC BY-NC-ND license（http://creativecommons.org/licenses/by-nc-nd/4.0/）.

1.Introduction

In recent years，there has been an increasing number of studies examining burnout among intensive care unit（ICU）nurses.1，2Despite extensive research focused on this topic，burnout continues to be challenging for critical care nurses worldwide3，and in China，burnout in the ICU is higher than in the general wards.3，4Burnout is a psychological syndrome in response to chronic emotional and interpersonal stressors on the job.It is contagious，and may cross over from one nurse to another in the ICU.5One study reported that nearly one-third of the ICU nursing team showed a high level of burnout.6Lederer et al.7maintained that advanced technology in the ICU can be a stressor，contributing to overall burnout；thus，nurses often feel overwhelmed and stressed when they first start working in the ICU.They will spend a long time in the ICU before feeling confident and competent，and they are often dependent on the expertise of experienced nurses.6A review of the literature from 2007 to 2012 revealed that nurse managers play a crucial role in preventing burnout by creating a supportive work environment for critical care nurses.2

In China，burnout is a significant issue for nurses，and Lin et al.4measured burnout in 249 randomly selected nurses from various wards of a large teaching hospital in Beijing.One hundred and twenty-eight nurses returned the completed questionnaire（response rate of 51%）.These results showed moderate levels of emotional exhaustion and personal accomplishment and low levels of depersonalisation.Xie et al.8proposed that nurses in Shanghai were suffering from high levels of burnout，which were strongly associated with work-related stress.The well-being of the nursing team is important for the quality of care，and a lack of social support was found to have a direct effect on emotional exhaustion and depersonalisation.9One study also showed that the best predictor of burnout appeared to be dissatisfaction with the emotional support received from supervisors.10

China is different from western countries in terms of social relationship patterns，nursing leadership models and administrative strategies.11Most hospitals still use a non-family system in which nurses rotate through a set roster.There are no casual or on-call nurses for sick leave and holiday relief，and consequently，nursing turnover and absenteeism can be high.12Support in such a situation is very relevant.Research on how social support relates to burnout among ICU nurses has not been performed in the ICU wards of large teaching hospitals in Shanghai，China.Thus，our study focused on ICU nurses with social support from others and on the effect of social support on burnout.

2.Material and Methods

A cross-sectional design was used to investigate the social support and burnout in ICU nurses.We selected five of the Grade III general hospitals in Shanghai with random cluster sampling.Participants in this study were voluntary and anonymous，and ethical approval for this study was provided by the Institutional Review Board of the Institute of Psychology of the University to which the authors are affiliated.The inclusion criteria included qualified nurses who had worked more than one year in the ICU.After permission was obtained from the hospital director，an appointment with the nurse-in-charge was made.Researchers distributed the questionnaires in the ICU wards of the hospital，and an appointment was made to collect the completed questionnaires after one week.A total number of 396 questionnaires were distributed to the nurses in five public hospitals in Shanghai.

The present study examined ICU nurse burnout and the informative，emotional，instrumental and appreciative support received from the supervisors，co-workers，friends and families as well as the correlation between the levels of burnout and social support.

Data were collected using self-reported questionnaires under the instruction of trained investigators in September 2012.A total of 362 nurses completed the questionnaires，providing a response rate of 91.4%.After a list-wise deletion of the cases with missing values，the final number of respondents was 356.The age of the subjects ranged from 20 to 47 years，with a mean age of 26.96（SD 4.07）.The majority of the sample group was unmarried（59%），with only a small percentage of the nurses being divorced or widowed（2%）.The average number of working years was 5.39（SD 4.59），ranging from 1 to 26 years，and the highest number of working years was not more than 5（64.6%）.The number of years of education in nursing varied between 3 and 6 years，but only a few of the nurses（1.4%）had a master's level of education.

Measures：Socio-demographic data were obtained by self-report on a form designed specifically for this study. The form included personal demographic questions about birth date，gender，education and income level，marital status，and race/ethnicity.

Background information：Personal details were obtained about the participants'job titles，gender，age，marital status，length of employment in the ICU ward，number of years of education for nursing，time for learning professional knowledge，and psychological and communications training.

Burnout：A translated Maslach Burnout Inventory-Human Services Survey-Chinese Version（MBI-HSSCV）13was used to examine the level of burnout among Chinese nurses.The survey tool is a 22-item questionnaire with three subscales：nine items for Emotional Exhaustion（EE），five items for Depersonalisation（DP），and eight items for personal accomplishment（PA）.The reliability coefficients for the subscales reported by Maslach and Jackson（1996）were 0.90 for EE，0.79 for DP，and 0.71 for PA，and the MBI-HSS was professionally translated into Chinese by Pang et al.13in Hong Kong，following the questionnaire back-translation processes（English-Chinese-English）.The reliability of the MBI-HSS-CV was reported as 0.773 7 for Chinese nurses.We received permission to use the translated MBI-HSS-CV from Pang et al.13Each item was answered on a seven-point response scale，scored from 0 to 6 as‘never'‘a few times a year'‘once a month or less'‘a few times a month'‘once a week'‘a few times a week'and‘every day'respectively.The responses were combined to obtain separate scores for each of the three subscales，which could then be categorised as low，average or high degrees of burnout according to the normative data.14

Social support：This scale was based on the“Social Support Rating Scale，”which was compiled by Xiao Shui Yuan.While social support can be provided by four main sources，that is，families，friends，work colleagues and the immediate supervisor，15every source should include the following four items：

Informational—reports can be obtained from colleagues on a critical matter.

Emotional—providing care，love，and trust.

Instrumental—providing facilitation behaviours to help the person meet work tasks.

Appraisal—obtaining evaluation and feedback on one's performance from the immediate supervisor.

The subscales used 5-point Likert scales，from 1（never）to 5（many）.This survey tested the reliability and validity of the scale，with the split-half reliability being 0.891 and 0.915（or better）.The structural validity evaluation of Cronbach's coefficient was above 0.9，and the correlation coefficient was greater than the total score and each factor.Thus，we propose that the scale scores have internal consistency.

Data analyses：Descriptive statistics were performed to examine the level of burnout，and the completed questionnaires were scored using the scoring key developed by Maslach and Jackson.16The final scores consisted of three sub-scores：EE，DP，and PA.Higher scores of EE and DP and lower scores of PA indicate a high level of burnout.

Profile analyses were performed to determine the differences in the four supporters.Multiple regression analyses were used to examine the predictive effect factors on burnout.Stepwise variable selection methods（probability of F，using 0.05 for entry and 0.10 for removal）were used to generate the multiple linear regression models.The variables considered for inclusion in each model included the participants'job titles，gender，age，marriage state，length of employment in the ICU ward，number of years of education in nursing，psychological and communication training and social support（family，friends，work colleagues and immediate supervisor）.These factors were related to the burnout subscale and were entered into the regression model.All analyses were performed using SPSS 16.0.The significance level used in this analysis was P＜0.05.

3.Results

3.1.Descriptive statistics

This survey included 356 nurses of Shanghai ICU，with an average age of 26.96±4.07 years（median 26）（Table 1）.73.7%of nurses who attended at least one year of psychological training，and 41.4%of nurses with nearly three years without training.Nurses who participated in interpersonal communication training accounted for 31.5%，and approximately one-third of the nurses did not take leave.

Table 1 Socio-demographic data of ICU nurses in Shanghai.

Approximately 61.7%of the nurses did not contribute to academic exchanges，and those willing to engage with other workers accounted for 54.9%of the participants.36.0%of nurses who did not want to work in their current areas，and 75.6%of nurses think that job requirements is too high.

3.2.Levels of burnout

In this study，the burnout reports ranged from 5 to 54 for emotional exhaustion（M=31.85，SD=8.38），0 to 30 for depersonalisation（M=11.69，SD=5.54），and 2 to 48 for personal accomplishment（M=19.79，SD=7.02）.These results showed that the nurses'levels of EE，DP and PA were in the high level category of burnout，according to Maslach's cut-off point scale（Maslach，Jackson，1996）.The high rates of EE，DP and PA were 76.4%，39.6%and 94.9%，respectively（Table 2）.

Table 2 Prevalence of risk levels of types of burnout

3.3.Profile analysis of social support

A profile analysis of social support indicated that nurses received sufficient support from their family but less support from their co-workers and friends.However，they received minimum support from their supervisors.There was no parallel，cross or overlap among the supervisor support，family support，co-worker support or friend support（Fig.1）.

3.4.Multiple regression analysis（MRA）

The normality test results of the analysis were as follows：n=356；emotional exhaustion u=0.066，P=0.066；depersonalisation u=0.102，P=0.102；low personal accomplishment u=0.047，P=0.047.These results showed three dependent variable values for the approximate normal distribution（or Gaussian distribution）. There were 17 independent variables in the questionnaire for multiple factors regression analysis to extract the main factors.

3.4.1.Emotional exhaustion

Four variables were selected for the single factor regression analysis.We applied multiple stepwise linear regression analysis to these four factors（such as social supper with immediate supervisor and co-workers），which indicated a statistically significant difference（R2= 0.111，F=10.719）.The results of the multiple linear regression analysis are shown in Table 3.

Fig.1.Profile analysis of social support.

Table 3 Multiple regression analysis of emotional exhaustion.

As shown in Table 3，The major influencing factors of emotional exhaustion included support from co-workers（b=0.343，t=1.980，P=0.049），taking leave（b= -1.182，t=-3.747，P=0.001），requisition of work（b= -1.41，t=-2.369，P=0.018），and supervisor support（b=-0.524，t=-3.926，P=0.001）.Taking leave and requisition of work had significantly negative correlation with emotional exhaustion.Co-worker support had a significantly positive correlation with emotional exhaustion.

Lower emotional exhaustion was reflected by incorporating expertise，additional information，emotional，instrumental and appraisal support from supervisors，longer leave time and a higher requisition of work.The more the informational，emotional，instrumental and appraisal support was obtained from coworkers，the lower was the emotional exhaustion.

3.4.2.Depersonalisation

Four variables were selected for single factor regression analysis.We applied multiple step-wise linear regression analysis to the immediate supervisor，age and requisition of work，indicating a statistically significant difference（R2=0.073，F=9.101）.The effect of the multiple linear regression analysis results is shown in Table 4.

Table 4 Multiple regression analysis of depersonalisation.

As shown in Table 4，the major influencing factors of depersonalisation were support from the supervisor（b= -0.333，t=-4.146，P=0.001），age（b=-0.89，t= -2.272，P=0.024）and requisition of work（b=-0.148，t=-2.124，P=0.034）.There was a negative correlation between depersonalisation and supervisor support，age and requisition of work.The more was the informational，emotional，instrumental and appraisal support from supervisors，the lower was the depersonalization exhaustion.The older the age and the higher the requisition of work，the lower was the depersonalization exhaustion.

3.4.3.Personal accomplishment

This variable was selected for single factor regression analysis.We applied the multiple stepwise linear regression analysis，and the immediate supervisor indicated a statistically significant difference（t=-2.096，P=0.037）with the regression equation of low personal accomplishment.

The major influencing factor of low personal accomplishment was support from the supervisor.The superior give more information or advice and more concern，understanding and support，the stronger nurse's professional accomplishment，the less prone to job burnout.

4.Discussion

The mean scores of emotional exhaustion and personal accomplishment indicated that Shanghai ICU nurses experienced high levels of burnout，while results of depersonalisation indicated that the group experienced a moderate level of burnout.These findings revealed that 272 of 356 ICU nurses exhibited severe emotional exhaustion，while 94.9%of nurses had a high degree of low personal accomplishment.This may be due to the particularly high risk of developing burnout by nurses who work in the intensive care unit（ICU）because of their experience of chronic stress.5，7

In a study by Bakker et al.5burnout was shown to be contagious among nurses working in ICUs.Among the nurses who reported the highest prevalence of burnout，their colleagues were most likely to experience high levels of burnout themselves.These findings indicate that interventions are needed to address the burnout level of nurses in the ICU in Shanghai，China.

4.1.Supervisor and co-worker support contribute to emotional exhaustion

This study indicated that there is a lower level of emotional exhaustion with more informational，emotional，instrumental and appraisal support from supervisors. Supervisory support had a significant negative relationship with emotional exhaustion.However，co-worker support had a significant positive relationship with emotional exhaustion.According to the profile analysis of social support，we found that support from the immediate supervisor is in a weak position.The results of this investigation indicated high emotional exhaustion.Prins et al.10reported a high score rate（76.4%）and that the best predictor of burnout was dissatisfaction with the emotional support received from supervisors.

Multiple stepwise linear regression analyses revealed that taking leave and requisition of work had a significantly negative relationship with emotional exhaustion. The longer the time for taking leave and the higher the requisition of work，the lower is the emotional exhaustion.Emotional exhaustion is not a symptom of work stress，but it is the end result of unmanaged work stress. When workloads are too heavy and demands are too great，then nursing care suffers since ideals may clash from reality.

4.2.Supervisor and family support affecting depersonalisation

The results of this study showed that there is a nega-tive correlation between depersonalisation and supervisory support；however，there is a positive correlation between depersonalisation and family support.According to the profile analysis of social support，we found that support from an immediate supervisor is in a weak position and that family support is in a strong position.The setting of the ICU determines the difficulty in communication and feelings of isolation，as well as feelings of conflict or aggressiveness，which are very common in Shanghai.This contributes to depersonalisation exhaustion.The resulting disappointment and failed personal expectations are a breeding ground for burnout in nurses.It is a debilitating psychological condition brought about by unrelieved work stress，which results in depleted energy reserves，lowered resistance to illness，increased dissatisfaction and pessimism，increased absenteeism and inefficiency at work.

Other research shows that a lack of social support affects burnout and stress symptoms1.Leiter et al.17reported a research on a nursing work life model of burnout identified emotional exhaustion as a contributing factor to depersonalisation，which is understood as a coping，self-preserving technique in response to emotional overload.18When critical care nurses are unable to resolve moral conflict，such as in unnecessarily prolonging life，a common response is to place distance between themselves and the situation.19Gutierrez likens this tendency to the basic“flight or fight”response to a threat.

With older age，increased frequencies of night shifts and longer professional knowledge learning time，depersonalisation exhaustion becomes lower.Older nurses have more experience，higher professional knowledge，and a stronger ability to judge and solve problems.For many younger nurses who have fallen into the first generation of the“one child per family”Chinese national policy，parents and grandparents will usually take on most of the responsibility for daily living activities.Consequently，they may have fewer responsibilities in their lives than their counterparts.Older nurses（particularly married nurses）usually take on more responsibilities than younger nurses because ICU patients are in critical condition and critical care nurses have the closest contact with them.At the same time，critically ill patients have great demands for complex care.2，7

4.3.Supervisor support determines low personal accomplishment

In this study，the major influencing factor of low personal accomplishment was support from the supervisor.There was a negative correlation between low personal accomplishment and supervisory support.Many critical care nurses have such high standards of care and expectations of themselves that they are not able to meet them and，consequently，are at risk of feeling a lack of personal accomplishment.20These nurses can also feel a lack of effectiveness in morally distressing situations that they are unable to resolve.19Feelings of ineffectiveness are further compounded when the critical care nurse does not feel supported by his or her nurse manager.Nurses working in an intensive care unit rarely see patients returning home when hospital treatments are successful. Thus，in this situation，being unable to see outcomes of patients they care for，may lead to limited feelings of success.

4.4.Limitations and strengths

This study has some limitations.It was based on a cross-sectional design with secondary analysis of the baseline data from a larger predictive Burnout Emotion（BE）study.Consequently，no conclusions could be drawn about the temporal sequence in the association between demographic factors and BE.Factors that may influence BE were limited to the variables selected for the larger study，and this restricted the explanatory possibilities for understanding the variance in BE.

In addition，although the rate of the responses to the questionnaire was excellent and was most likely representative，we still had a rate of 8.6%non-respondents. We cannot exclude the possibility that these collaborators did not complete the questionnaire because they had a high degree of burnout and were unable to provide any additional effort.

In the multivariate analysis，due to the relatively small sample size，we included only variables with P＜0.05 in the univariate analysis.In doing so，we could have excluded some relevant variables.Nevertheless，we cannot conclude that an increase in our mortality rate could be associated with an even higher level of burnout.

This study also had notable strengths.

The high participation rate indicated that the ICU nursing team had confidence in the investigators and was willing to express their opinions.

5.Conclusions

The results obtained from this study indicated that there is a high level of burnout exhaustion in ICU nurses in Shanghai，China.Our findings suggested that supervisor support，co-worker support，taking leave and requisition of work are major influencing factors of emotional exhaustion.The major factors influencing depersonalisation were support from the supervisor，age and requisition of work.More support from the supervisor may be an important strategy to improve personal accomplishment.Future research and interventions should be directed beyond the individual level to broader social and environmental influences.

Conflicts of interest

All contributing authors declare no conflicts of interest.

References

1.Verdon M，Merlani P，Perneger T，Ricou B.Burnout in a surgical ICU team.Intensive Care Med.2008；34：152-156.

2.Epp K.Burnout in critical care nurses：a literature review. Dynamics.2012；23：25-31.

3.McClendon H，Buckner EB.Distressing situations in the critical care unit.Dimensions Crit Care Nurs.2007；26：199-206.

4.Lin F，St John W，Mcveigh C.Burnout among hospital nurses in China.J Nurs Manag.2009；17：294-301.

5.Bakker AB，Le Blanc PM，Schaufeli WB.Burnout contagion among intensive care nurses.J Adv Nurs.2005；51：276-287.

6.McGrath M.The challenges of caring in a technological environment：critical care nurses'experiences.J Clin Nurs.2008；17：1096-1104.

7.Lederer W，Kinzl JF，Traweger C，Dosch J，Sumann G.Fully developed burnout and burnout risk in critical care personnel at a university hospital.Anaesth Crit Care.2008；36：208-213.

8.Xie Z，Wang A，Chen B.Nurse burnout and its association with occupational stress in a cross-sectional study in Shanghai. J Adv Nurs.2011；67：1537-1546.

9.Bakker AB，Demerouti E，Verbeke W.Using the job demands-resources model to predict burnout and performance，Hum Resour Manag.2004；43：83-104.

10.Prins JT，Hoekstra-Weebers JEHM，Gazendam-Donofrio SM，et al.The role of social support in burnout among Dutch medical residents.Psychol Health Med.2007；12：1-6.

11.Smith DR，Tang S.Nursing in China：historical development，current issues and future challenges.Int J Nurs Stud. 2004；5：16-20.

12.Hu J，Liu H.Job satisfaction among nurses in China.Home Health Care Manag Pract.2004；17：9-13.

13.Pang MCS，Chin CF，Choi K，et al.Burnout in front line nurses in caring of the terminally ill.Hongkong；Hospital Authority，Residents；1997.

14.Maslach C，Schaufeli WB，Leiter MP.Job burnout.Ann Rev Psychol.2001；52：397-422.

15.Ganster DC，Fusilier MR，Mayes BT.Role of social support in the experience of stress at work.J Appl Psychol.1986；71：102-110.

16.Maslach C，Jackson SE.Maslach burnout inventory.Manual.Palo Alto，CA：Consulting Psychologists Press Inc：1996.

17.Leiter MP，Spence Laschinger HK.Relationships of work and practice environment to professional burnout：testing a causal model.Nurs Res.2006；55：137-146.

18.Bühler K，Land T.Burnout and personality in critical care：an empirical study.Hosp Topics.2003；81：5-12.

19.Gutierrez KM.Critical care nurses'perceptions of and responses to moral distress.Dimensions Crit Care Nurs.2005；24：229-241.

20.Stayt LC.Nurses'experiences of caring for families with relatives in critical care units.J Adv Nurs.2007；57：623-630.

23 December 2014

in revised form

.

E-mail address：sunbook_2001@126.com（H.Ruan）.

Peer review under responsibility of Shanxi Medical Periodical Press.

http：//dx.doi.org/10.1016/j.cnre.2015.04.003

2095-7718/ⓒ2015 Shanxi Medical Periodical Press.Production and hosting by Elsevier B.V.This is an open access article under the CC BY-NC-ND license（http：//creativecommons.org/licenses/by-nc-nd/4.0/）.