Network analysis of the relationships among burnout, presenteeism, and social support in Chinese pediatric nurses.

Background: Burnout and presenteeism are occupational health problems affecting professional stability in the nursing workforce and the quality of healthcare services. Social support is considered to be associated with burnout and presenteeism; however, the symptom-level association patterns among these three constructs remain unclear. This study systematically explored the structural associations and node characteristics among burnout, presenteeism, and social support based on a network analysis approach. Methods: A questionnaire survey was conducted among pediatric nurses in Chongqing, Southwest China, using a convenience sampling method, yielding 876 valid samples. A Gaussian graphical model (GGM) was constructed based on the Spearman correlation matrix, and the regularized partial correlation network was estimated using the EBIC-glasso algorithm. Network structure was computed using R software, and centrality indices, including strength, betweenness, closeness, and expected influence were derived. The accuracy of edge weight and the stability of centrality measures were assessed through 1,000 nonparametric bootstrap samples. Results: Presenteeism was positively correlated with burnout (r = 0.433), social support was negatively correlated with burnout (r = -0.361), and social support was negatively correlated with presenteeism (r = -0.179) (all p < 0.05). The strongest edge weight in the network structure was 0.72. Within the burnout module, D6 showed a strong connection with D5 and was associated with the depersonalization item E1, indicating that emotional exhaustion occupied a core position in the network. Centrality analysis demonstrated that E1 (Strength = 1.173), C11 (Strength = 1.122), and D6 (Strength = 1.112) were influential nodes. D1 and C3 showed the highest betweenness (Bet. = 97), playing bridging roles. Bootstrap results indicated narrow confidence intervals for edge weights, and the strength index maintained correlations >0.70 when 50% of the sample was retained, suggesting acceptable stability for strength-related findings in this sample. Conclusions: There were interactive relationships among burnout, presenteeism, and social support in pediatric nurses. Items related to emotional exhaustion were relevant nodes linking depersonalization and presenteeism, while family and emergency-related support resources played roles within the social support module. Network analysis identified relevant intervention nodes in occupational health and provides a reference for promoting occupational health.