Complaints of daytime sleepiness, insomnia, hypnotic use, and risk of dementia: a prospective cohort study in the elderly

Table 4 Associations between sleep complaints, hypnotics use, and incidence of dementia with vascular component over a 12-year follow-up

Variables	Whole sample		Incident DVC
	n = 6295		No, n = 6156		Yes, n = 139		Model 1^a		Model 2^b		Model 3^c
	n	%	n	%	n	%	HR [95%CI]^d	p	HR [95%CI]^d	p	HR [95%CI]^d	p
Excessive daytime sleepiness (EDS)								0.001		0.008		0.02
No	5186	82.38	5090	82.68	96	69.06	1		1		1
Yes	1109	17.62	1066	17.32	43	30.94	1.87 [1.28–2.74]		1.68 [1.15–2.47]		1.58 [1.07–2.32]
Difficulty with initiating sleep (DIS)								0.52		0.63		0.99
No	4098	65.10	4018	65.27	80	57.55	1		1		1
Yes	2197	34.90	2138	34.73	59	42.45	1.13 [0.78–1.62]		1.09 [0.76–1.58]		1.00 [0.69–1.45]
Difficulty in maintaining sleep (DMS)								0.62		0.98		0.78
No	2290	36.38	2247	36.50	43	30.94	1		1		1
Yes	4005	63.62	3909	63.50	96	69.06	1.10 [0.76–1.58]		1.01 [0.69–1.45]		0.95 [0.66–1.37]
Early morning awakening (EMA)								0.64		0.56		0.38
No	4006	63.64	3921	63.69	85	61.15	1		1		1
Yes	2289	36.36	2235	36.31	54	38.85	0.92 [0.64–1.31]		0.90 [0.63–1.29]		0.85 [0.60–1.22]
Sleep quality (SQ)								0.37		0.27		0.08
Good	2692	47.53	2630	47.46	62	50.41	1		1		1
Bad	2972	52.47	2911	52.54	61	49.59	0.85 [0.59–1.21]		0.82 [0.57–1.17]		0.72 [0.50–1.04]
Clinical severity sleep^e (CSS)	6.18 (±3.22)		6.17 (±3.22)		6.53 (±3.05)		0.99 [0.94–1.05]	0.81	0.98 [0.92–1.04]	0.46	0.96 [0.90–1.02]	0.15
Clinical severity sleep (CSS)								0.06		0.09		0.06
[0; 4]	1922	33.93	1893	34.16	29	23.58	1		1		1
[4; 7.5]	2019	35.65	1961	35.39	58	47.15	1.58 [1.00–2.48]		1.41 [0.89–2.22]		1.30 [0.83–2.06]
> 7.5	1723	30.42	1687	30.45	36	29.27	1.03 [0.62–1.72]		0.91 [0.55–1.52]		0.79 [0.47–1.33]
Hypnotic use								0.02		0.04		0.26
No	4929	78.30	4833	78.51	96	69.06	1		1		1
Yes	1366	21.70	1323	21.49	43	30.94	1.52 [1.06; 2.18]		1.46 [1.01–2.11]		1.24 [0.85–1.81]

^aAdjusted for study center, sex, mobility, and presence of the APOE-ε4 allele, stratified for the level of education
^bAdjusted for all covariates in model 1 plus diabetes mellitus, body-mass index, and cardiovascular disease
^cAdjusted for all covariates in model 2 plus depressive status
^dEstimated using Cox proportional hazard models with delayed entry and age as the time scale
^eContinuous variables were expressed as mean (standard deviation)

ISSN: 1758-9193