Independent risk factors for the development of incontinence-associated dermatitis (category 2) in critically ill patients with fecal incontinence: A cross-sectional observational study in 48 ICU units

Abstract

Background: Critically ill patients suffering from fecal incontinence have a major risk of developing incontinence-associated dermatitis (IAD). The presence of moisture and digestive enzymes (lipase, protease) negatively influences skin barrier function. Additional risk factors will make some patients even more vulnerable than others. In order to provide (cost) effective prevention, this specific patient population should be identified timely.

Objectives: To identify independent risk factors for the development of IAD category 2 (skin loss) in critically ill patients with fecal incontinence.

Design: A cross-sectional observational study.

Setting and participants: The study was performed in 48 ICU wards from 27 Belgian hospitals. Patients of 18 years or older, with fecal incontinence at the moment of data collection, were eligible to participate. Patients with persistent skin redness due to incontinence (IAD category 1) were excluded.

Methods: Potential risk factors were carefully determined based on literature and expert consultations. Data were collected over a period of eight months by trained researchers using patient records and observation of skin care practices. At the time a patient was included in the study, all relevant data from the past six days, or since admission at the ICU, were recorded. Simultaneously, direct skin observations were performed and high definition photographs were ratified by an expert IAD researcher. A multiple binary logistic regression model was composed to identify independent risk factors. Variables with P < .25 in single binary logistic regression analyses were added to the multiple model using a forward procedure. A cut-off value of P < .1 was established to retain variables in the final model. Nagelkerke's R² and Hosmer-Lemeshow statistic were calculated as measures of model fit.

Results: The sample comprised of 206 patients, of which 95 presented with IAD category 2, and 111 were free of IAD. Seven independent risk factors were identified: liquid stool [odds ratio (OR) 4.69; 95% confidence interval (CI) 2.28-9.62], diabetes (OR 2.89; 95% CI 1.34-6.27), age (OR 1.05; 95% CI 1.02-1.08), smoking (OR 2.67; 95% CI 1.21-5.91), non-use of diapers (OR 2.97; 95% CI 1.39-6.33), fever (OR 2.60; 95% CI 1.23-5.53), and low oxygen saturation (OR 2.15; 95% CI 1.03-4.48). Nagelkerke's R² was 0.377. The Hosmer-Lemeshow statistic indicated no significant difference between the observed and expected values (p = .301).

Conclusions: Liquid stool, diabetes, age, smoking, non-use of diapers, fever, and low oxygen saturation were independently associated with IAD category 2 in critically ill patients with fecal incontinence.

Keywords: Critical illness; Fecal incontinence; Incontinence-associated dermatitis; Intensive care units; Nursing; Perineal dermatitis; Predictor; Risk.