Observational study of predictive monitoring to spot critical illness before symptoms appear

CHAOS is based on the investigator's new and exciting results from pre-clinical and large longitudinal multi-center observational clinical studies of critically ill patients and asymptomatic community-based adults with little or no advanced disease.

By integrating approaches from the physical, biological, computational, statistical and clinical sciences, this observational study will test the hypothesis that early diagnosis of subclinical signatures of critical illness encoded within physiological signals complements conventional clinical predictors by providing unique prognostic insight.

The primary goal is to reduce mortality, morbidity and complications by early identification of individuals with brewing subclinical critical illness and adverse events before overt clinical presentation (e.g., cardiac arrest, arrhythmias, hemorrhage, respiratory failure, circulatory collapse). This will provide the necessary lead time for healthcare providers to deliver early, more effective and/or preventive therapies. Through innovative approaches, CHAOS also meets the challenge of medical errors to reduce missed diagnosis, misdiagnosis, preventable harm and variability in provider adherence to best practice guidelines.

The goal is to validate predictive algorithms and identify subclinical signatures of illness, ranging from asymptomatic adults in the community to very sick patients in the hospital. The overall goal is to make healthcare more precise, effective, efficient, safe and timely while reducing costs, preventable harms and adverse events.