Underdiagnosis and undertreatment is a major problem in childhood asthma management, especially in preschool-aged children. Current prognostic approaches using risk-score based tools have poor-to-modest accuracy, are impractical, and have limited evidence of efficacy in clinical settings and hence are not widely used in practice. The objective of the study is to determine the usability, acceptability, feasibility, and preliminary efficacy of the childhood asthma passive digital marker (PDM) among pediatricians. The study will include practicing pediatricians within the IU Health Network.
Study Type
INTERVENTIONAL
Allocation
RANDOMIZED
Purpose
SCREENING
Masking
SINGLE
Enrollment
34
A childhood asthma Passive Digital Marker (PDM) is an ML algorithm that is able to retrieve and synthesize pre-existing "passively" collected mother/child dyad prognostic data in "digital" electronic health record (EHR) to provide an objective and quantifiable "marker" of a child's risk (probability) and associated pathophysiological phenotype to inform clinician decision-making at point-of-care.
Indiana University
Indianapolis, Indiana, United States
Perceived PDM Acceptance
Mean Perceived PDM acceptance measured using a Behavioral Intention scale (BIS) with a score Likert scale score of 0-5. Here, higher score values represent a higher acceptability of the PDM (i.e., better outcome).
Time frame: 8 to 12 months
Perceived PDM Usability
Mean Perceived Usability measured using a modified Simplified System Usability Scale (SUS) with a score Likert scale of 0-5. Here, higher score values represent a higher perceived usability of the PDM (i.e., better outcome).
Time frame: 8 to 12 months
Study Feasibility
Percent of successful study enrollment of eligible clinicians (\>80%)
Time frame: 8 to 12 months
Prognostic Accuracy
Clinician prognostic accuracy was defined as the proportion of correctly classified vignettes. Accuracy was calculated as: correct vignette classifications ÷ total vignettes evaluated. Values ranged from 0 to 1, with higher values indicating greater prognostic accuracy. Each clinician assessed 10 vignettes (5 cases and 5 controls) and classified them as either high or low risk. A correct classification was high risk for a case and low risk for a control vignette.
Time frame: 3 to 12 months
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.