| Objective:1.To investigate the current situation and explore the relationship between psychological capital and burnout among oncology nurses.3.To provide a basis for prevention and intervention by establishing and verifying a prediction model of burnout in oncology nurses.Methods:1.From march to August,2018,1 224 oncology nurses from Guangxi were selected to conduct questionnaires on general demographic,psychological capital questionnaire,Gallup Workplace Audit,Maslach Burnout InvestigationGeneral Survey,nursing ability scale and turnover intention scale.The differences of burnout among different demographic data were analyzed by independent sample t test or one-way ANOVA.The influence factors of burnout were analyzed by means of multiple linear regression.Pearson correlation analysis was used to analyze the correlation between variables.Structural equation model was further used to analyze the relationship between psychological capital and burnout.2.The Random forest prediction model was constructed using the data of oncology nurses in Nanning.The verification of the model selects nurse from hospitals outside Nanning.The selected variables with the greatest impact on burnout of oncology nurses and the general demographic of oncology nurses were taken as input,and the occurrence of burnout was taken as output.The accuracy was evaluated using the area under the ROC curve.Results:1.The average score of burnout among oncology nurses was(2.45±0.79),with 84.5% of oncology nurses in some burnout and 9.2% in severe burnout;turnover intention was(2.53±0.95),which was at a medium level;The score of psychological capital(4.15±0.87)was in the middle level,which was at a medium level;and the engagement score(33.64±5.25)was high.The score of nursing ability scale(73.44±13.11)was in the middle level.2.Multiple linear regression analysis showed that psychological capital,income satisfaction,night shift,patient number cased each day,rotary department,the satisfaction of the relationship between nurses and patients,working years of oncology nurses,age is the independence of oncology nurse job burnout influence factors.Psychological capital including is the most important factors of oncology nurses(P < 0.05).3.Correlation analysis showed that burnout of oncology nurses was negatively correlated with psychological capital,engagement and nursing ability(r=-0.253,-0.276,-0.076,P<0.05),positively correlated with resignation intention(r=0.437,P<0.05);Further analysis of the structural equation model showed that psychological capital of oncology nurses had a negative predictive effect on burnout(β=-0.19,P < 0.05),and engagement played a mediating role(β=-0.16,P < 0.05).4.The area under the ROC curve of the model were 0.764,indicating that the model had a good prediction effect.A total of 10 explanatory variables were selected,which were satisfaction with income,satisfaction with nurse-patient relationship,working years of oncology nurses,optimism,psychological capital,age,the number of training per year,recognition and respect of public,night shift,and hope.Conclusion:1.It is not optimistic that burnout and turnover intention of oncology nurses.Psychological capital and nursing ability among oncology nurses need to be improved.Engagement level of oncology nurses is high.2.Oncology nurses’ burnout which was affected nursing ability and turnover intention should be paid attention.Burnout of oncology nurses is significantly related to psychological capital,which can indirectly affect burnout of oncology nurses through engagement.3.Oncology nurses’ burnout are influenced by many factors,psychological capital,the satisfaction of income,night shift,the number of patients cared with each day,rotary department,satisfaction with nurse-patient relationship,working years in oncology nurses,age is the independence of oncology nurse’burnout influence factors.Nursing managers should relief burnout by creating supportive working environment of the development of psychological capital.4.The burnout prediction model based on psychological capital and individual characteristics of oncology nurses has a good prediction effect.In the future,it can be considered to build an early warning platform for burnout of oncology nurses and formulate targeted psychological intervention strategies to prevent and intervene burnout of oncology nurses. |