Risk factors for not achieving stoma self-care proficiency at hospital discharge and development of a clinical prediction model: A study based on the Danish Stoma Database

Aim: Stoma self-care is critical for long-term adaptation. Personalised approaches may help identify patients needing additional guidance during initial admission. This study aimed to examine factors associated with not achieving stoma self-care proficiency and develop a risk model for use in clinical practice. Method: We linked prospective health data from the Danish Stoma Database Capital Region and the Danish Anaesthesia Database. Multiple logistic regression analysis provided risk factors associated with not achieving stoma self-care proficiency. The C-statistic was used to evaluate the discriminative ability of logistic regression models. Data were randomly split into training and validation sets to develop models and estimate risk in predefined groups. Results: Of 5826 pre-determined for ostomy self-care, 4910 (84%) achieved proficiency at discharge, while 916 (16%) did not. Patients of older age, female sex, smokers, higher BMI and ASA scores were more likely not to perform stoma self-care. Significant treatment factors were the type of surgery, surgical approach and ‘stoma not visible to patient’. The risk model included five dichotomous factors: age ≥ 50, age ≥ 70, ASA score ≥ III, acute surgery and female sex. Each risk factor provided a score of 0 or 1 point, resulting in a maximum of five points for high-risk patients. The C-index was 0.66. Conclusion: Baseline characteristics, treatment factors and stoma visibility were associated with a reduced likelihood of achieving stoma self-care proficiency. The risk model, which incorporates readily known variables, may help clinicians identify patients at risk. The model should be further tested in a clinical setting.