Predictors of asthma control in severe paediatric asthma

<bold>Aim:</bold> To identify predictors of achieving asthma control in severe pediatric asthma. <bold>Method:</bold> This research was conducted within the Paediatric Asthma Non-invasive Diagnostic Approaches (PANDA) study, a prospective observational cohort involving children 6-17 years, treated according to GINA 2018 guideline step ≥3. Information on current asthma medication, exacerbations, comorbidities, exposure to triggers, and lung function parameters was collected from medical records. Additionally, questionnaires on asthma symptoms and adherence were administered ((Childhood) Asthma Control Test ((C-)ACT), Medication Adherence Report Scale 5). Asthma control was defined as a (C-)ACT score of ≥20 points and no asthma exacerbations in the preceding 6 months. Trough bivariate analyses we identified potential predictors of asthma control, and created a prediction model using binary logistic regression. <bold>Results:</bold> 69 patients (mean age 11.8±3.1 years, 65% male) with a mean follow-up time of 7.0±1.4 months, were included. 61% had uncontrolled asthma at inclusion compared to 51% at study completion. A (C-)ACT score <20 points and a higher number of asthma exacerbations in the 6 months prior to inclusion were significantly (p <0.05) associated with uncontrolled asthma at study completion. The predictive model including (C-)ACT score (OR 3.7, 95% CI 1.3-10.5, p=.015) and exacerbations (OR 1.5, 95% CI 0.9-2.4, p=.101) had an accuracy of 68%, sensitivity of 69%, and area under the ROC-curve of 0.72. <bold>Conclusion:</bold> This study underscores the complexity of achieving good asthma control in children with severe asthma and the need for new non-invasive markers to improve identification of patients with persistently high disease burden.