Risk-Profile Based Monitoring Intervals for Multivariate Longitudinal Biomarker Measurements and Competing Events...

Area of Science:

• Biomedical Engineering
• Clinical Informatics
• Cardiology

Background:

• Routine patient monitoring is crucial for managing chronic conditions and detecting disease exacerbations.
• Current monitoring programs often use fixed schedules, which may not be optimal for individual patient needs.
• Chronic heart failure (CHF) management requires regular assessment of disease progression.

Purpose of the Study:

• To propose and evaluate an adaptive scheduling strategy for patient monitoring.
• To improve the efficiency of monitoring programs for chronic conditions like CHF.
• To leverage dynamic individual risk predictions for personalized patient management.

Main Methods:

• Developed an adaptive scheduling strategy based on dynamic individual risk predictions.
• Incorporated multiple longitudinal measurements and competing events into the strategy.
• Conducted a simulation study using data from the Bio-SHiFT cohort of stable CHF patients.
• Compared the adaptive strategy against fixed schedule alternatives.

Main Results:

• The adaptive strategy demonstrated potential for improved efficiency in monitoring programs.
• Dynamic risk predictions, using patient characteristics and biomarkers (e.g., NT-proBNP, troponin), informed the scheduling.
• Simulation results indicated advantages over fixed monitoring schedules for stable CHF patients.

Conclusions:

• An adaptive scheduling strategy based on dynamic risk prediction can enhance the efficiency of chronic disease monitoring.
• This approach is particularly relevant for conditions like stable chronic heart failure (CHF).
• Personalized monitoring schedules can optimize resource allocation and patient care.