Feasibility and acceptability of a passive digital marker for childhood asthma risk prognosis: a pilot cross-sectional study
摘要
Electronic health records (EHR)-based childhood asthma prediction may be a valid, cost-effective, and scalable option for early detection of at-risk children for targeted preventive intervention, but this approach has not been adopted in clinical practice. The purpose of this pilot study was to evaluate the feasibility and acceptability of the Pediatric Asthma Risk Score (PARS) as an EHR-based passive digital marker (PDM) for childhood asthma prognosis in clinical practice.
MethodsFeasibility was defined by the level of concordance between the PARS and a clinician’s classification of a case study’s risk of developing school-age asthma. Acceptability was based on clinicians’ intentions to use the PARS and its usability as a PDM for childhood asthma prognosis. Blinded to the PARS risk classification, clinicians were asked to predict the case study’s risk of school-age asthma based on their professional judgment as either low or high risk and rate their confidence in their prediction. Unblinded to the PARS risk classification, clinicians rated the acceptability of the PARS as a PDM. Logistic regression models were used to summarize and identify correlates of feasibility and acceptability measures.
ResultsThere was 74% (95% CI: 66–81%) concordance between the PARS and clinician prediction of our case study’s risk of developing school-age asthma. Risk discordance (26%) was associated with low confidence in a clinician’s professional judgment (adjusted odds ratio [aOR]: 4.78; 95% CI: 1.85, 12.34) and PARS (aOR: 4.48; 95% CI: 1.16, 17.33). Unblinded to the PARS risk classification, more than 80% of clinicians ranked their intentions to use the PARS and its usability as a PDM as 4+ on a 0–6 Likert scale.
ConclusionOur findings demonstrate preliminary feasibility and acceptability of the Pediatric Asthma Risk Score (PARS) as a passive digital marker for childhood asthma risk, supporting further evaluation in prospective real-world studies.