Table 2 Sleep parameter differences (minutes) between estimates from sleep diary and two accelerometer-based methods (N = 25,645 nights, N = 3752 individuals).

From: Estimating sleep parameters using an accelerometer without sleep diary

Sleep parameters

HDCZA

L5 ± 6

Method

Sleep onset time

Waking time

SPT-window duration

Sleep onset time

Waking time

SPT-window duration

Y-intercept (SE)

−12.5 (0.9)**

−1.6 (0.8)P = 0.04

10.9 (1.1)**

−86.5 (1.0)**

45.9 (0.9)**

131.7 (1.2)**

Betas (SE)

Women

5.0 (1.1)**

−3.0 (0.9)*

−8.0 (1.3)**

8.0 (1.4)**

−8.6 (1.1)**

−16.2(1.6) **

Ten years of age

3.9 (0.8)**

−2.9 (0.7)**

−6.8 (1.0)**

0.3 (1.0) P = 0.78

0.2 (0.8) P = 0.83

−0.2 (1.2) P = 0.89

Five BMI index points

1.0 (0.5) P = 0.06

−1.5 (0.5)*

−2.5 (0.7)**

−3.2 (0.7)**

1.8 (0.6)*

4.8 (0.8)**

Weekend

3.0 (1.0)*

2.0 (0.9) P = 0.02

−1.0 (1.2) P = 0.41

6.4 (1.3)**

−0.3 (1.0) P = 0.77

−6.3 (1.4)**

Winter

1.0 (0.9) P = 0.27

−1.2 (0.8) P = 0.12

−2.2 (1.1) P = 0.05

−1.0 (1.2) P = 0.39

0.6 (1.0) P = 0.51

1.7 (1.3) P = 0.2

Within individual residual SD

24.7

21.3

30.9

18

13

20.6

Between individual residual SD

66.1

56.9

82.3

88.9

74.9

101.8

AIC

81175

73538

92433

94053

84956

100978

  1. Degrees of freedom = 25,645; Relative to mean age of 69.1 years; Relative to mean BMI of 26.4 kg/m2; SE: Standard Error; SD: Standard Deviation; AIC = Akaike information coefficient, *P < 0.005, **P < 0.0005.