iPREDICT: proof-of-concept study to develop a predictive model of changes in asthma control.

Area of Science:

• Digital health technologies
• Asthma management
• Predictive modeling

Background:

• The individualized PREdiction of DIsease Control using digital sensor Technology (iPREDICT) program utilizes digital tools for asthma management.
• Devices are integrated into patients' daily lives to monitor health status changes with minimal burden.

Purpose of the Study:

• Establish baseline asthma characteristics.
• Detect health status changes associated with asthma events.
• Evaluate algorithms for trigger identification and prediction of asthma control fluctuations.
• Explore patient experience and device compliance.

Main Methods:

• A 24-week, multicenter, observational study in the US.
• Included patients aged 12+ with severe, uncontrolled asthma.
• Utilized spirometers, vital sign monitors, sleep monitors, connected inhalers, and mobile apps with patient-reported outcome questionnaires.
• Linked prospective data with electronic health records for algorithm development.
• Defined asthma events based on symptoms, peak expiratory flow (PEF), forced expiratory volume in 1s (FEV1), and short-acting beta-agonist (SABA) use.

Main Results:

• Out of 108 patients, 66 completed the study.
• Predictive accuracy varied by endpoint; population-level models showed low accuracy for PEF <65%.
• Subgroup models demonstrated high accuracy for specific allergies, triggers, and asthma types.
• The most accurate predictive endpoint was >4 SABA puffs/day/48h.
• Individual models showed significant accuracy for PEF <65% and >4 SABA puffs/day/48h.

Conclusions:

• Multidimensional datasets enabled population, subgroup, and individual analyses.
• Provided proof-of-concept for developing predictive models of fluctuating asthma control.
• Highlights the potential of digital health technology in personalized asthma management.