Skip to main content
. 2023 Jul 20;4(1):zpad028. doi: 10.1093/sleepadvances/zpad028

Table 4.

PSG and Sleep Disorder Predictors of the Difference Between GGIR Actigraphy TST Without a Sleep Diary and PSG TST (HDCZA Actigraphy-PSG, Hours)

Univariable Multivariable
Characteristic Beta (95% CI)a P-value Beta (95% CI)a P-value
(Intercept) −0.8 (−0.88, −0.72) <0.001
Actigraphy TST (mean-centered), hours 0.64 (0.60, 0.69) <0.001 0.56 (0.52, 0.60) <0.001
PSG SOL, hours 0.89 (0.66,1.1) <0.001 0.79 (0.68, 0.91) <0.001
PSG WASO, hours 0.56 (0.45, 0.66) <0.001 0.81 (0.76, 0.87) <0.001
AHI, per 10 events/hour −0.08 (−0.12, −0.03) 0.001
Arousal index, events/hour −0.01 (−0.01, 0.00) 0.007
Insomnia symptoms −0.34 (−0.57, −0.12) 0.003
Daytime sleepiness (ESS) −0.03 (−0.05, −0.01) 0.005
PSG 3% Oxygen desaturation index, desaturations/hour −0.01 (−0.02, 0.00) <0.001

aCI, Confidence Interval.

Multivariable model: R2/R2 adjusted = 0.78/ 0.78.

Variables were not in the multivariable model due to nonsignificance (AHI, insomnia, p > 0.05) or multicollinearity (sleep efficiency, arousal index, and 3% oxygen desaturation index).