Understanding nurses’ challenging work experiences: a text mining-based sentiment analysis and topic modeling study

Nurses experience substantial psychological burden during clinical practice. Previous studies have provided limited insights into the emotional dimensions of these experiences and lack specificity in identifying task-level contributors to nursing work experiences. To identify and categorize challenging nursing work experiences using a systematic text-based approach. This study analyzed a subset of data from the SMART NURSE project, comprising 1,650 free-text responses from nurses working in two general hospitals in South Korea. Sentiment analysis was conducted to quantify emotional polarity using a lexicon-based approach. Keyword analysis and Latent Dirichlet Allocation topic modeling were applied to identify the key terms and underlying themes. Topic prevalence and C v coherence were assessed to evaluate the model adequacy. Overall, the sentiment was skewed toward negative emotions. The most frequently mentioned word was “response.” Topic modeling revealed three distinct themes: concurrent task demands, patient flow, and communication load. Concurrent task demands showed the highest prevalence and C v coherence. The most challenging nursing work experiences are characterized by the interplay of concurrent task demands, patient flow, and communication-related burdens. Integrating emotional and contextual dimensions through text mining provides a more comprehensive understanding of nursing work experiences and supports the development of targeted workload management strategies and context-sensitive interventions.