Prevalence of and characteristics associated with insomnia and obstructive sleep apnea among veterans with knee and hip osteoarthritis

Table 2 Bivariate and Multivariable Linear Regression Models of Insomnia (ISI) Scores

Bivariate and Multivariable Linear Regression Models of Insomnia (ISI) Scores
	Bivariate Model			Multivariable Model
	Estimate (SE)	95% CI	p	Estimate (SE)	95% CI	p
Age	−0.29 (0.05)	− 0.38, − 0.19	<.001	−0.09 (0.04)	− 0.16, − 0.02	.01
Gender
Male	1.34 (1.59)	−1.78, 4.46	.40	−0.20 (1.07)	−2.30, 1.89	.85
Race
Non-Hispanic White	−3.07 (0.91)	−4.86, −1.28	.001	−0.19 (0.65)	−1.47, 1.08	.76
Self-Rated Health
Excellent, Very Good, or Good	−5.82 (0.89)	−7.58, −4.07	<.001	−0.82 (0.73)	−2.26, 0.62	.26
BMI	0.11 (0.08)	−0.04, 0.27	.15	−0.06 (0.06)	−0.17, 0.06	.34
PTSD Diagnosis	5.65 (0.98)	3.73, 7.57	<.001	1.68 (0.73)	0.25, 3.11	.02
WOMAC Pain	1.05 (0.10)	0.85, 1.25	<.001	0.36 (0.10)	0.17, 0.55	<.001
Depressive Symptoms
PHQ-8	1.10 (0.06)	0.99, 1.22	.001	0.84 (0.07)	0.70, 0.98	<.001
# Joints with Arthritis	0.49 (0.12)	0.25, 0.74	< 0.001	0.07 (0.09)	−0.11, 0.25	.44
Opioid Use	3.38 (0.99)	1.42, 5.33	< 0.001	0.18 (0.71)	−1.23, 1.58	0.81

All bivariate models had 2 missing observations. Otherwise, the unadjusted model with the model with WOMAC Pain has 1 additional missing observation and the model with PHQ-8 and the model with opioid use have 4 additional missing observations. The multivariable model has 11 missing observations. Negative scores associated with less insomnia symptoms. Higher WOMAC scores indicate greater pain. Self-rated health was recoded into a binary variable: excellent, very good, and good health vs. fair/poor health

ISSN: 1471-2474