Table 4.

Univariate and multivariate logistic regression analysis for the highest quintile (Q5) of winter morning surge in systolic blood pressure

	Univariate regression analysis			Multivariate regression analysis
	Odds ratio	95% CI	P value	Odds ratio	95% CI	P value
Age ≥75 y	3.36	1.89‐5.97	<0.001	3.73	2.00‐6.95	<0.001
Male	0.91	0.54‐1.54	0.722	0.89	0.37‐2.12	0.793
Body mass index ≥25 kg/m2	0.55	0.33‐0.91	0.021	0.65	0.38‐1.12	0.118
Current smoking	0.77	0.29‐2.08	0.608	0.80	0.25‐2.64	0.719
Daily drinker	0.84	0.44‐1.59	0.588	1.12	0.43‐2.93	0.823
Preexisting CHD	0.31	0.04‐2.37	0.257	0.21	0.03‐1.68	0.141
Preexisting stroke	2.31	0.94‐5.65	0.067	2.14	0.76‐6.00	0.148
Dyslipidemia	1.05	0.63‐1.75	0.855	1.12	0.61‐2.06	0.717
Diabetes mellitus	1.07	0.59‐1.94	0.833	0.83	0.42‐1.66	0.597
Increase in antihypertensives	0.84	0.52‐1.36	0.476	0.82	0.48‐1.41	0.470
Decrease in antihypertensives	2.86	0.99‐8.29	0.053	3.02	0.90‐10.12	0.073
Occupant‐owned housing	2.26	1.20‐4.29	0.012	2.58	1.29‐5.16	0.007

Preexisting CHD includes preexisting angina pectoris and myocardial infarction. Multivariate regression analysis was adjusted for age ≥75 y, sex (male), body mass index ≥25.0 kg/m², current smoking, daily drinking, history of CHD, history of stroke, dyslipidemia, diabetes mellitus, increase in antihypertensives, decrease in antihypertensives, and occupant‐owned housing.

CHD, coronary heart disease; CI, confidence interval.