Table 5.

Linear associations between sleep quality scores and covariates.

(A) Univariate linear regression models between sleep quality scores and DAS28-CRP as a continuous variable.

	ESS			AIS			PSQI
	Estimate	p	Model R²	Estimate	p	Model R²	Estimate	p	Model R²
DAS28	Relationship not statistically significant			0.898	< 0.0001	0.0769	0.69459	< 0.0001	0.0580

(B) Univariate logistic regression models between sleep quality scores and DAS28-CRP as a continuous variable.

	ESS		AIS		PSQI
	OR (95% CI)	p	OR (95% CI)	p	OR (95% CI)	p
DAS28	Relationship not statistically significant		1.300 (1.104–1.530)	0.0016	1.409 (1.173–1.692)	0.0002

(C) Multivariate linear regression models by sleep scale with DAS28-CRP as a continuous variable.

ESS	Estimate	p	AIS	Estimate	p	PSQI	Estimate	p
	R² = 0.235			R² = 0.437			R² = 0.391
	p < 0.0001			p < 0.0001			p < 0.0001
	n = 264			n = 286			n = 274
DAS28	−0.507	0.0248	DAS28	0.478	0.0034	DAS28	0.336	0.0269
PANAS Score N	0.144	0.0084	Concurrent medication: glucocorticoids	0.976	0.0417	Sleep medication	3.245	< 0.0001
SF-36 General Health	0.220	0.0213	Comorbidities	1.588	0.0152	Living with partner	−1.017	0.0376
SF-36 Mental Health	0.238	0.0499	PANAS Score N	0.224	< 0.0001	PANAS Score N	0.182	< 0.0001
SF-36 Vitality	−0.283	0.0478	PANAS Score P	−0.096	0.0028	PANAS Score P	–0.069	0.0250

(D) Multivariate logistic regression models by sleep scale with DAS28-CRP as a continuous variable.

ESS	OR (95% CI)	p	AIS	OR (95% CI)	p	PSQI	OR (95% CI)	p
	n = 261			n = 286			n = 270
	LR < 0.0001			LR < 0.0001			LR < 0.0001
	Score < 0.0001			Score < 0.0001			Score < 0.0001
	Wald = 0.0021			Wald = 0.0001			Wald = 0.0003
DAS28	2.388 (1.142–4.993)	0.0207	DAS28	1.229 (1.002–1.508)	0.0483	DAS28	1.348 (1.058–1.718)	0.0157
PANAS Score N	1.105 (1.035–1.180)	0.0029	Comorbidities	2.324 (1.023–5.277)	0.0439	Sleep medication	16.741 (3.607–77.700)	0.0003
SF-36 General Health	1.154 (1.025–1.298)	0.0174	Caffeinated drinks	2.297 (1.017–5.183)	0.0453	PANAS Score N	1.106 (1.049–1.166)	0.0002
SF-36 Mental Health	1.205 (1.034–1.403)	0.0167	PANAS Score N	1.097 (1.050–1.146)	< 0.001
SF-36 Role Physical	0.865 (0.759–0.984)	0.0280	PANAS Score P	0.961 (0.924–0.999)	0.0466
SF-36 Vitality	0.813 (0.679–0.972)	0.0234

For multivariate linear and logistic regression models, only factors found to be significantly associated with sleep quality are presented. For multivariate linear regression models: covariate estimate > 0: increase of the covariate leads to an increase of sleep quality score (worsens sleepiness); covariate estimate < 0: increase of the covariate leads to a decrease of sleep quality score (improves sleepiness). For multivariate logistic regression models: OR < 1: means that the variable reduces the risk of having poor sleep quality and denotes a negative relationship of the covariate with the risk of having excessive daytime sleepiness; OR > 1: means that the variable increases the risk of having poor sleep quality and denotes a positive relationship of the covariate with the risk of having excessive daytime sleepiness. AIS: Athens Insomnia Scale; DAS28-CRP: 28-joint Disease Activity Score (C-reactive protein); ESS: Epworth Sleepiness Scale; LR: likelihood ratio; PANAS P and N: Positive and Negative Affect Schedule positive and negative emotions; PSQI: Pittsburgh Sleep Quality Index; SF-36: Medical Outcomes Study Short Form-36 Health Survey.