Table 5: Multivariate analysis (linear regression) of the possible predictors (sociodemographic, clinical variables and context) of sleep-wake patterns with total sleep time, number of wakes, wake in percent, and minutes sleep at daytime as dependent variables (standardized beta, 𝑁 = 1 0 6 ).

Total sleep time (TST)WakesWake % of TSTDaytime sleep
𝛽 𝑃 𝛽 𝑃 𝛽 𝑃 𝛽 𝑃

Step 1. Sociodemographic
 Age 0.19.18 0.15.30−0.03.85−0.02.99
 Sex (men as ref.) 0.21.047−0.13.21−0.11.31 0.11.31
 Education (low, high) 0.05.60 0.03.77−0.02.88 0.03.75
 Paired relationship (paired as ref.)−0.09.41−0.12.24−0.06.56 0.08.44
 Work (paid work as ref.)−0.05.74 0.04.78 0.01.96 0.07.60

 Explained variance ( 𝑅 2 ) 7.2% 5.7%0.7% 3.6%

Step 2. Clinical
 Physical functioning (SF-36–PF) 0.02.89−0.06.56−0.08.47−0.08.49
 Sedatives 0.00.99−0.01.95 0.14.19−0.08.47
 Time since admission (days) 0.05.63−0.05.64−0.11.30 0.19.08

 Explained variance ( 𝑅 2 ) 7.5% 7.3% 4.3% 8.5%
𝑅 2 change 0.3% 1.6% 3.7% 4.9%

Step 3. Context
 Slept at home all nights
 (hospital as ref.)
0.10.38−0.32.007−0.21.09−0.02.84
 Slept both at hospital and home
  (hospital as ref.)
0.13.25−0.10.39−0.06.58−0.07.56

 Explained variance ( 𝑅 2 ) 8.9% 14.3% 7.2% 8.9%
𝑅 2 change 1.4% 7.0% 2.9% 0.3%