Validity, reliability and discriminative capacity of an electronic quality of life instrument (Pelican) for childhood asthma in the Netherlands

Abstract

To assess psychometric properties of the Pelican instrument, an online Dutch self-administered Quality of Life instrument for childhood asthma for scientific and clinical use. A cohort study was done in two asthma populations and healthy children. One asthma population had assessment at start, 4 and 8weeks. The other asthma population and healthy children had one assessment. All children were aged 6-12years. Children completed the Pelican instrument, Paediatric Asthma Quality of Life Questionnaire, feeling thermometer and Childhood Asthma Control Test. Lung function and fraction exhaled nitric oxide were measured. Parents completed Functional Status II, Asthma Control Questionnaire, Childhood Asthma Control Test questionnaires and symptom diaries. We assessed interpretability, structural validity, internal consistency, reliability, construct and discriminative validity of the Pelican instrument. Eighty-five asthmatic (mean age 8.5years) and 49 healthy children (mean age 8.4years) participated. The Pelican instrument has 5 domains with 21 items after factor analysis. Internal consistency was 0.89 (CI 0.85-0.92), domain reliability showed Cronbach's α's from 0.64 to 0.76 and item-to-scale correlations from 0.61 to 0.81. Test-retest reliability was confirmed ICC=0.88 (CI 0.79-0.93). Construct validity was demonstrated by significant moderate correlations with other relevant asthma outcomes like PAQLQ (r=-0.59, p<0.01). Discriminative capacity between controlled or uncontrolled asthma (t=3.20, p<0.01, Δ=0.64) and asthma versus healthy subjects (t=6.31, p<0.01, Δ=0.94) was found. The psychometric properties of the Pelican instrument were acceptable in Dutch paediatric asthma patients between 6 and 12years old.