Construction and internal validation of an exploratory risk-prediction model for missed nursing care among neuro-oncology nurses: a multicenter cross-sectional study

Missed nursing care (MNC) occurs frequently in neuro-oncology departments and severely threatens patient safety. This study aimed to explore influencing factors of MNC among neuro-oncology nurses and construct a nomogram prediction model for clinical risk assessment. A total of 638 neuro-oncology nurses from 12 Grade A (tertiary-tier) cancer hospitals across seven Chinese provinces and municipalities were enrolled in this multicenter cross-sectional survey. A general information questionnaire and the Oncology MNC Self-rating Scale, Workplace Violence Scale, Professional Identification Scale, and Clinical Decision-making in Nursing Scale were adopted for data collection. Univariate analysis and multivariate logistic regression were used to screen independent risk factors of MNC. A nomogram was established, and model performance was evaluated via receiver operating characteristic curve, calibration curve, and decision curve analyses; Hosmer–Lemeshow testing; and 1000 bootstrap resampling for internal validation. 433 participants (67.9%) reported at least one missed care item. Multivariate analysis revealed that female sex, long single-shift hours, frequent night shifts, long working years, supervisor nurse title, experience with workplace violence, low job satisfaction, poor professional identity, and insufficient clinical decision-making ability were independent risk factors for MNC. The established model presented good discriminative power, with an original area under the curve (AUC) of 0.910 and an optimism-corrected AUC of 0.899. The calibration curve fit well with the ideal 45° line, and the Hosmer–Lemeshow test indicated good model calibration ( P = 0.424). Decision curve analysis confirmed favorable clinical utility of the model. The nomogram model shows robust internal predictive performance and clinical practicability, yet external validation with independent samples is still required to improve its generalizability across diverse medical institutions. It can help nursing managers identify nurses at high risk of MNC early and support targeted interventions. Further external validation is needed to improve its generalizability. This study provides a scientific reference for reducing MNC and promoting nursing quality and patient safety in neuro-oncology settings. Not applicable.