Abstract: Objective: To systematically evaluate risk prediction models for Incontinence-Associated Dermatitis (IAD), providing references for the development of such models and clinical practice. Methods: A comprehensive search was conducted in Chinese and English databases for literature on risk prediction models for IAD, with the search period spanning from database inception to March 17, 2025. Two researchers independently performed literature screening, data extraction, and assessment of the methodological quality of the included studies using the Prediction Model Risk of Bias Assessment Tool (PROBAST). Results: Seven studies were included, comprising a total of eight IAD risk prediction models. The Area Under the Curve (AUC) for the models ranged from 0.810 to 0.904, with sensitivity of 60.9%–99.4% and specificity of 79.41%–96.7%. Most models were presented as formula-based. The most common predictive factors included: age [OR=1.52, 95%CI (1.16, 2.00)], serum albumin level [OR=2.35, 95%CI (1.75, 3.16)], and frequency of bowel movements [OR=3.21, 95%CI (2.45, 4.20)]. Conclusion: Research on IAD risk prediction models is still in the developmental stage. Future studies should refine design and reporting protocols, validate existing models, and further evaluate their clinical applicability and effectiveness to facilitate translation into practice. This will enable personalized interventions to reduce IAD incidence.

Key words: Incontinence-Associated Dermatitis; risk prediction model; Meta-analysis