Psychometric evaluation of an electronic Asthma Symptom Diary for young children

Abstract

BACKGROUND: Patient-reported outcome measures that facilitate self-report by children are needed to reduce the bias of proxy report. We previously developed an electronic Pediatric Asthma Symptom Diary (ePASD) to assess the severity of daily asthma symptoms and proximal impacts in children aged 6-11 years with mild to severe asthma. The ePASD, administered via a digital application with visuals, sounds, and text, is uniquely designed to minimize the importance of reading skills on children's ability to self-report accurately. Here, we describe the ePASD's psychometric properties.

METHODS: Ninety-one children aged 6-11 years with mild to severe asthma and their caregivers participated in 2 study visits, which consisted of training on the provisioned device and completing asthma-specific clinical outcome assessment (COA) questionnaires. The children self-completed the ePASD at home twice daily for 8 consecutive days. The scoring of the ePASD was guided by factor analyses, inter-item correlations, and internal consistencies. Reliability, discriminating ability, construct validity, and responsiveness were evaluated for ePASD items and candidate scores.

RESULTS: All COAs included in the study-the ePASD, Asthma Control Questionnaire (ACQ), Childhood Asthma Control Test, Pediatric Asthma Quality of Life Questionnaire-Standardized (PAQLQ[S]), and global ratings-demonstrated that the children exhibited few asthma-related symptoms and impacts at all timepoints, and consequently, showed little change over time. Internal consistencies (all Cronbach's alphas ≥ 0.52) and test-retest reliabilities (all intraclass correlation coefficients ≥ 0.60) were largely satisfactory. Patterns of convergent and divergent correlations supported the construct validity of ePASD scores. The ePASD symptom scores correlated moderately to strongly with PAQLQ(S) Symptom scores (all correlations ≥ - 0.46) and with ACQ scores (all correlations ≥ 0.42), as predicted. Evidence of the discriminating ability of ePASD items and composite scores was demonstrated by known-groups analyses.

CONCLUSIONS: The ePASD is a reliable and valid measure of asthma symptoms and proximal impacts in children aged 6-11 years with mild, moderate, or severe asthma. These results lay the psychometric groundwork for use of the ePASD in future clinical trials for the management of pediatric asthma. An ongoing pediatric asthma treatment trial is anticipated to provide evidence of the ePASD's responsiveness to change.