Objective To investigate the incidence of incontinence‐associated dermatitis (IAD) among patients in the intensive care unit (ICU) and to identify potential risk factors to establish a reference for clinical nursing work. Background Patients in the ICU are susceptible to IAD. IAD is painful, reduces the patient's quality of life and adds to the workload of clinical medical staff. However, risk factors associated with IAD may differ between countries and healthcare settings. Design Prospective cohort study Methods From November 2016 to November 2017, a prospective cohort study was conducted among109 patients in three Class 3, Grade A hospitals (comprising 9 ICUs in total) in Beijing. The Incontinence‐associated Dermatitis and Its Severity (IADS) instrument in Chinese was applied to assess IAD. Univariate and multivariate logistic regression analyses were performed to identify risk factors for IAD. Results The study population had 29 community‐acquired and 80 nosocomial infections, and 26 (incidence: 23.9%) of these had IAD. On univariate analysis, a significant difference was observed between patients with and without IAD with respect to the following indices: Barthel index, Braden scale score, Nutritional Risk Screening 2002 (NRS2002) score, serum albumin level, occurrence of infection, faecal incontinence, frequency of faecal incontinence, stool property and double (faecal and urinary) incontinence and perineal assessment tool (PAT) score (p < 0.05). Multivariate logistic regression analysis showed that three factors entered the regression equation—that is, the Braden Scale Score, serum albumin level and double incontinence. Of these, the Braden Scale Score and serum albumin level were protective factors for IAD. Thus, the higher the Braden Scale Score, the lower the risk of IAD (OR = 0.678, 95% confidence interval CI = 0.494–0.931); a higher level of serum albumin implies a lower risk of IAD, provided it is within the normal range (OR = 0.884, 95%CI = 0.797–0.981). Double incontinence was an independent risk factor for IAD (OR = 10.512, 95% CI = 2.492–44.342). Conclusion A higher morbidity of IAD is seen in the ICU. Specific preventive and nursing measures are required to maintain the skin integrity of critically ill patients in daily nursing practice to improve patient quality of life and the quality of nursing care. Relevance to clinical practice Incontinence‐associated dermatitis is characterised by inflammation and tissue damage due to prolonged/repeated exposure to urine and/or stool. Not every patient with urine and/or stool incontinence develops IAD. Medical staff can use research‐based evidence to identify ICU patients at risk of IAD to reduce morbidity and improve health outcomes.