JMIR Publications

ABSTRACT

Background:

Burnout (BO), a critical syndrome and particularly for nurses in healthcare settings, substantially affects their physical and psychological status, the institute’s well-being, and indirectly patient outcomes. However, objectively classifying burnout levels has not been definitely defined and noticed in the literature.

Objective:

The aim of this study is to build a model using the convolutional neural network (CNN) to develop an app for automatic detection and classification of nurse BO using the Maslach BO Inventory—Human Services Survey (MBI—HSS) for helping assess nurse BO at an earlier stage.

Methods:

We recruited 1003 nurses working in a Taiwan medical center to fill out the Chinese version of the 20-item MBI—HSS in August 2016. The k-mean and the CNN were used as unsupervised and supervised learnings, respectively, for (1) dividing nurses into two classes (n = 531 and 472 of suspicious BO+ and BO−, respectively) and (2) building a BO prediction model to estimate 38 parameters. Three additional featured variables (i.e., the mean, standard deviation (SD), and Lz person-fit statistic) were added to the 20-item model (called the 23-item model) to improve prediction accuracy. Finally, data were separated into training and testing sets in a proportion (70%:30%)—the former was used to predict the latter. We calculated the sensitivity, specificity, and receiver operating characteristic curve [area under the curve (AUC)] across studies for comparison. An app predicting the respondent’s BO was developed involving the model’s 38 estimated parameters for a website assessment.

Results:

We observed that (1) the 23-item model yields a higher accuracy rate (0.98) with an AUC (0.98; 95% CI 0.97–1.00) higher than that of the 20-item model with an accuracy of 0.95 and an AUC of 0.97 (95% CI; 0.86–0.99), based on the 1003 cases; (2) the scheme named matching personal response to adapt for the correct classification in model drives the model’s accuracy at 100%; (3) the 700-case training set with 0.96 accuracy predicts the 303-case testing set reaching an accuracy of 0.91; and (4) an available MBI—HSS app for nurses predicting BO was successfully developed and demonstrated in this study.

Conclusions:

The 20-item model combined with the three featured variables (i.e., the response mean, SD, and Lz) is recommended to estimate the parameters in the CNN for improving the accuracy of nurse BOs. An app developed for helping nurses’ self-assess job BO at an early stage is required for application in the future. Clinical Trial: Not available