Gender differences in the bidirectional relationship between alcohol consumption and sleeplessness: the Tromsø study

Table 3 Logistic regression analysis of sleeplessness for women according to change in alcohol use

Alcohol use	Model 1^a		Model 2^b		Model 3^c		Model 4^d
Alcohol use	OR	CI _.95	OR	CI _.95	OR	CI _.95	OR	CI _.95
Stable high, high T1-high T2	0.92	0.56–1.49	1.11	0.67–1.85	1.12	0.65–1.94	1.11	0.64–1.92
Increasing, low T1-high T2	1.02	0.75–1.39	1.31	0.95–1.80	1.24	0.88–1.75	1.28	0.90–1.81
Decreasing, high T1-low T2	1.07	0.79–1.46	1.12	0.82–1.53	1.07	0.76–1.52	1.05	0.74–1.49
Stable low, low T1-low T2	Ref.		Ref.		Ref.		Ref.

Notes. *** p < .001, ** p < .01, * p < .05. CI_.95 = 95% confidence interval
^aCrude estimates
^bAdjusted for age, education and marital status
^cAdjusted for Model 2 + sleeplessness at T1
^dAdjusted for Model 3 + mental health problems, somatic health problems, BMI, and social welfare benefits

ISSN: 1471-2458