Endpoints and Measurements

Wrist actigraphy demonstrates agreement rates of 91–93% against polysomnography, with sensitivity above 91% for detecting the sleep period time window. Seven days of continuous accelerometer wear provides more reliable sleep estimates than the one to two nights typically recorded during polysomnography. Heart-rate tracking is described as accurate within one beat per minute compared with a medical-grade EKG. The system measures heart rate, respiration, and movement during sleep. The system combines heart rate, respiration, and movement to infer sleep timing, sleep stages, stage percentages, and total sleep duration. The mattress cover uses embedded sensors and ballistocardiography to detect body signals during sleep. Actigraphy measures inactivity as a proxy for sleep and cannot distinguish true sleep from motionless wakefulness. Subjective sleep questionnaires commonly used in heart failure research require recall periods of 2–4 weeks and carry substantial recall bias risk, particularly in older adults. Actigraphy tends to overestimate sleep duration and underestimate wake after sleep onset relative to polysomnography, suggesting true sleep health may be worse than the reported f…