Digital physiological biomarkers predict within-person symptom changes in complex chronic illness.

Area of Science:

• Digital Health
• Cardiovascular Physiology
• Chronic Disease Management

Background:

• Altered heart rate variability (HRV) and resting heart rate (HR) are prevalent in complex chronic conditions.
• Mobile and wearable technologies offer real-time, validated measurements of HRV and HR for symptom monitoring.
• These technologies are advancing the management of chronic illness symptoms.

Purpose of the Study:

• To investigate if within-person fluctuations in HR, HRV, and respiratory rate predict daily symptom changes (crash, fatigue, brain fog).
• To evaluate the predictive performance of mobile health data for symptom exacerbations.
• To assess the added value of morning biometrics for real-time symptom prediction.

Main Methods:

• Integration of 60-second morning photoplethysmography (PPG) assessments with evening symptom severity reports.
• Analysis of a high-density mobile health dataset (n=4244) with an average of 125 biometric observations per participant.
• Utilized walk-forward cross-validation to assess model predictive performance.

Main Results:

• Models incorporating morning biometrics alongside prior-day symptom reports showed the highest fit and variance explained.
• Within-person increases in HR and decreases in HRV in the morning correlated with worsening evening symptom reports.
• Adding morning biometrics to prior-day symptom reports significantly improved model performance (AUC=0.82-0.85 vs. 0.73-0.83).

Conclusions:

• Mobile health tools demonstrate prospective utility for precision monitoring in complex chronic illnesses.
• Real-time prediction of symptom exacerbations is feasible using integrated biometric and self-reported data.
• This approach supports personalized management strategies for individuals with chronic conditions.