Skip to main content

Table 3 Association of the Pittsburgh sleep quality assessment index (PSQI) with the variables of interest

From: Metabolic syndrome is associated with better quality of sleep in the oldest old: results from the “Mugello Study”

 

Linear regression model

PSQI as a continuous variable

B

95% CI

P

Age (each year)

− .05

− .20 to .09

.467

Sex (female)

.22

− .77 to 1.22

.658

Benzodiazepines

1.44

.26 to 2.61

.017

Platelet antiaggregants

− 1.84

− 2.76 to − .92

.000

ACE-inhibitors

− .14

− 1.25 to .97

.800

Beta-blockers

.27

− 1.02 to 1.56

.683

Charlson comorbidity Index

.07

− .13 to .27

.495

Heart failure

.92

− .16 to 2.00

.095

Albumin (proportion of 1) [%]

− .38

− 1.11 to .34

.296

Hemoglobin (g/L) [g/dL]

− .12

− .42 to .19

.443

Metabolic syndrome

− 1.04

− 2.06 to − .03

.044

 

Logistic regression model

PSQI < 5

OR

95% CI

P

Age (each year)

1.01

.91 to 1.12

.856

Sex (female)

1.03

.54 to 1.99

.920

Benzodiazepines

.31

.12 to .80

.015

Platelet antiaggregants

2.25

1.22 to 4.17

.010

ACE-inhibitors

.88

.47 to 1.67

.698

Beta-blockers

1.02

.46 to 2.29

.953

Charlson comorbidity Index

.92

.80 to 1.07

.276

Heart failure

.49

.23 to 1.08

.076

Albumin (proportion of 1) [%]

.91

.85 to .98

.012

Hemoglobin (g/L) [g/dL]

1.17

.95 to 1.44

.139

Metabolic syndrome

2.52

1.26 to 5.02

.009

 

Logistic regression model

PSQI ≤ 7

OR

95% CI

P

Age (each year)

.97

.89 to 1.06

.536

Sex (female)

.60

.33 to 1.06

.079

Benzodiazepines

.40

.20 to .81

.011

Platelet antiaggregants

1.85

1.08 to 3.16

.025

ACE-inhibitors

1.81

1.04 to 3.15

.037

Beta-blockers

.93

.44 to 1.94

.839

Charlson comorbidity Index

.95

.84 to 1.07

.424

Heart failure

.89

.47 to 1.66

.705

Albumin (proportion of 1) [%]

.95

.89 to 1.04

.069

Hemoglobin (g/L) [g/dL]

1.08

.90 to 1.30

.413

Metabolic syndrome

2.11

1.11 to 3.40

.022

  1. All the covariates were entered simultaneously into the regression models. The model included all the variables which differed significantly (P < .050) in univariable analyses in Tables 1, and 2