Figure 1. Study Timeline and Longitudinal Blood Pressure Patterns.

^aOf the participants who attended visit 4, 3304 died before visit 5 and 2247 did not attend visit 5. There were 433 participants who did not attend visit 4 but attended visit 5.

^bBetween visits 5 and 6, 1777 participants in the analytic sample were excluded (see panel C), 198 died, and 1284 did not attend visit 6.

^cBlood pressure pattern groups based on blood pressure levels at Visits 1 to 5. Dotted lines represent the approximate 24-year blood pressure pattern. Hypertension was defined as systolic blood pressure >140 mm Hg, or diastolic blood pressure >90 mm Hg, or use of antihypertensive medication. Hypotension was defined as systolic blood pressure <90 or diastolic blood pressure <60, irrespective of current antihypertensive medication use or hypertension diagnosis. Blood pressure patterns were defined using the standard hypertension definition. Midlife hypertension was defined as meeting hypertension criteria for two consecutive visits between Visits 1 and 4; persons not meeting this criterion were classified as midlife normotensive. Late-life normotension, hypertension, and hypotension were defined at Visit 5. Participants with midlife hypertension/late-life normotension were not included in the standard hypertension definition analyses because of small sample size (n = 23).

^dParticipants were excluded for missing information regarding the following: smoking status (n = 687), drinking status (n = 321), body mass index (n = 185), coronary heart disease (n = 100), total cholesterol (n = 85), total high-density lipoprotein cholesterol (n = 85), diabetes (n = 56), previous stroke (n = 10), and education (n = 9). A subset of participants were missing information for more than 1 covariate; these values are not mutually exclusive.

^eParticipants missing information (blood pressure or antihypertensive medication use) necessary to determine visit 5 hypertension status using the standard hypertension criteria were excluded.