A pilot validation study of contactless ballistocardiography versus polysomnography for sleep and respiratory assessment in older adults.

STUDY

OBJECTIVE

To evaluate the agreement between a contactless ballistocardiography (BCG) based sleep monitoring device and polysomnography (PSG) in assessing sleep macrostructure and respiratory parameters in older adults.

METHODS

Seventeen male participants (aged 50-80 years) underwent overnight PSG, while being simultaneously monitored using the BCG device. Parameters assessed included sleep latency, total sleep time (TST), durations of light sleep (N1 + N2), deep sleep (N3), rapid eye movement (REM) sleep, and derived non rapid eye movement (NREM) sleep, along with apnoea-hypopnea index (AHI). Paired t-tests and Bland Altman analysis were done to evaluate agreement. A P -value < 0.05 was considered statistically significant, with Bonferroni correction applied for multiple comparisons.

RESULTS

Significant differences were observed for TST, REM, deep sleep, wake time, and sleep latency. Dozee overestimated TST by + 83.00 min (95% LoA: -42.88 to + 208.7) and light sleep by + 39.88 min (LoA: -197.0 to + 117.2), while underestimating deep sleep (-29.53 min; LoA: -99.20 to + 40.14), REM sleep (-45.63 min; LoA: -137.7 to + 46.43), and time in bed (-69.41 min; LoA: -236.0 to + 97.21). Sleep latency was slightly underestimated (-8.94 min; LoA: -28.20 to + 10.32), and wake time after sleep onset was overestimated (+35.24 min; LoA: -37.21 to + 107.7). AHI was modestly overestimated by + 7.18 events/hour (LoA: -18.09 to + 32.44).

CONCLUSION

While the BCG based device shows potential for non-invasive, unobtrusive sleep monitoring, it demonstrated substantial discrepancies when compared to PSG. These limitations may compromise the clinical utility of such devices, especially in older adults with fragmented sleep or respiratory abnormalities. BCG based devices may offer value in large scale or home-based screening contexts if future algorithmic refinements improve its staging and respiratory event detection capabilities.