Table 4. Association between rise and fall of blood pressure and incident stroke, ischemic stroke, hemorrhagic stroke, and unspecified stroke.

Outcome	n/N	Tertile 1		Tertile 2 HR	Tertile 3
		HR (95% CI)	p-Value		HR (95% CI)	p-Value
Systolic blood pressure
Any stroke	971/9,958	1.27 (1.08–1.49)	<0.001	1 (ref)	1.12 (0.95–1.32)	0.17
Ischemic stroke	641/9,958	1.09 (0.90–1.33)	0.38	1 (ref)	0.99 (0.82–1.21)	0.95
Hemorrhagic stroke	89/9,958	1.39 (0.81–2.38)	0.24	1 (ref)	1.45 (0.85–2.47)	0.17
Unspecified stroke	241/9,958	1.88 (1.31–2.70)	<0.001	1 (ref)	1.50 (1.04–2.17)	0.03
Diastolic blood pressure
Any stroke	971/9,955	1.10 (0.94–1.28)	0.26	1 (ref)	0.88 (0.74–1.03)	0.11
Ischemic stroke	641/9,955	1.04 (0.86–1.26)	0.67	1 (ref)	0.82 (0.67–1.00)	0.06
Hemorrhagic stroke	89/9,955	0.79 (0.47–1.33)	0.38	1 (ref)	0.84 (0.51–1.40)	0.50
Unspecified stroke	241/9,955	1.43 (1.01–2.01)	0.04	1 (ref)	1.10 (0.77–1.58)	0.59

Tertiles 1, 2, and 3 for systolic blood pressure were <−0.4%/year, −0.4% to 2.1%/year, and >2.1%/year, respectively. Tertiles 1, 2, and 3 for diastolic blood pressure were <−0.8%/year, −0.8% to 2.0%/year, and >2.0%/year, respectively. Values in bold are statistically significant. Adjusted for age, sex, mean systolic or diastolic blood pressure, education, body mass index, alcohol consumption, smoking, cholesterol, high-density lipoprotein cholesterol, lipid-lowering medication use, blood-pressure-lowering medication use, anti-thrombotic medication use, type 2 diabetes mellitus, atrial fibrillation, prior coronary heart disease, and type of blood pressure measuring device. Standard deviation of variance of each tertile for systolic blood pressure: 0.04 (tertile 1), 0.03 (tertile 2), 0.03 (tertile 3). Standard deviation of variance of each tertile for diastolic blood pressure: 0.04 (tertile 1), 0.008 (tertile 2), 0.04 (tertile 3). n, number of participants with incident stroke; N, total study population; ref, reference.