Barriers to evidence-based nursing practice and proactive innovation behavior among thoracic surgery nurses in China: a latent profile analysis

Evidence-based nursing practice (EBNP) is essential in thoracic surgery nursing because of its high specialization and postoperative risks. However, existing research has paid limited attention to heterogeneity in barriers to EBNP among Chinese thoracic surgery nurses. In particular, evidence on the latent profiles of these barriers and their relationship with proactive innovation behavior (PIB) remains scarce. To identify the distinct latent profiles of barriers to EBNP among Chinese thoracic surgery nurses, examine associations between demographic factors and latent profile membership, and further explore the relationship between these latent profiles and PIB. A cross-sectional survey was conducted among 682 nurses from September to November 2025. Latent profile analysis (LPA) was used to identify profiles of barriers to EBNP, based on responses to the Barriers to Evidence-Based Nursing Practice Scale. Multinomial logistic regression analysis was performed to identify factors associated with latent profile membership. Hierarchical regression analysis was used to examine the association between barriers to EBNP and PIB. LPA identified two severity-based profiles of barriers to EBNP among nurses: a low-barrier profile (49.6%) and a high-barrier profile (50.4%). Married nurses (OR = 2.180, 95% CI: 1.351–3.517) and those with night shifts, including 1–4 night shifts/month (OR = 25.546, 95% CI: 6.047–107.920) and ≥ 5 night shifts/month (OR = 21.626, 95% CI: 5.065–92.340), had higher odds of belonging to the high-barrier profile, whereas nurses with a Master’s degree or above (OR = 0.282, 95% CI: 0.132–0.602) and those who had participated in EBNP training (OR = 0.355, 95% CI: 0.232–0.542) had lower odds. Different profiles of barriers to EBNP were associated with nurses’ PIB (Δ R 2 = 0.037, p < 0.001). Barriers to EBNP among thoracic surgery nurses were associated with PIB, although their additional explanatory contribution was modest, as indicated by the Δ R 2 of 0.037. The latent profiles mainly reflected differences in the overall severity of these barriers. Identifying different profiles may help nursing managers screen nurses who require targeted support and develop differentiated strategies to reduce barriers to EBNP and promote innovation. Not applicable.