Latent profile analysis and influencing factors of compassion fatigue in intensive care unit nurses

This study aims to examine the status and potential categories of compassion fatigue (CF) among intensive care unit (ICU) nurses and to identify its key influencing factors. This cross-sectional study recruited a total of 848 ICU nurses from 22 tertiary hospitals in Shandong and Hebei provinces using convenience sampling between September 2024 and February 2025. Data were collected using the Chinese version of the Compassion Fatigue Scale, the Perceived Social Support Scale, the Connor-Davidson Resilience Scale, and the Practice Environment Scale. Compassion fatigue data were obtained through electronic questionnaires. Data were analyzed using latent profile analysis, Pearson’s chi-square test, the Kruskal-Wallis H test, and multivariate logistic regression. ICU nurses had a median compassion fatigue score of 80.00 (74.00, 87.00), with 56% experiencing moderate to severe levels of compassion fatigue. Latent profile analysis identified three distinct categories: the Exhaustion-Vulnerable Profile (44.00%), the Intermediate-Adaptive Profile (29%), and the Resilient-Tolerant Profile (27%). Multivariate logistic regression analysis revealed that educational level, nature of employment, sleep hours per day, social support, practice environment, and psychological resilience were significant factors influencing classification into different latent compassion fatigue categories (all P < 0.05). ICU nurses exhibit three latent categories of compassion fatigue, reflecting heterogeneity within the population. Importantly, nursing administrators should implement humanistic and diversified management strategies to enhance nurses’ work enthusiasm and reduce compassion fatigue.