Figure 5.
Results of Cox regressions assessing the association between blood pressure phenotypes and incident cardiovascular disease amongst participants without prevalent cardiovascular disease at the third health check of the EPIC-Norfolk study. Model A—Univariable; Model B—Multivariable adjustment for age and ethnicity; Model C—Model B + body mass index, physical activity level, smoking and alcohol consumption; Model D—Model C + pre-existing co-morbidities (cardiovascular disease, diabetes mellitus, cancer, asthma, chronic obstructive pulmonary disease) and serum low-density lipoprotein cholesterol; Model E—Model D + antihypertensive treatment. SBP trajectories in men: Trajectory 1—borderline stable optimal SBP, Trajectory 2—rising normal/high normal SBP, Trajectory 3—rising grade 1 hypertension, Trajectory 4—grade 1 hypertension → high-normal SBP, Trajectory 5—grade 2 → grade 1 hypertension. SBP trajectories in women: Trajectory 1—stable optimal SBP, Trajectory 2—rising normal/high normal SBP, Trajectory 3—high-normal SBP → grade 1 hypertension, Trajectory 4—decreasing grade 1 hypertension, Trajectory 5—grade 1 → grade 2 hypertension. DBP trajectories: Trajectory 1—low optimal DBP, Trajectory 2—high optimal DBP, Trajectory 3—normal/high-normal DBP, Trajectory 4—grade 1 hypertension → normal DBP, Trajectory 5—stable grade 1/grade 2 hypertension. Median (interquartile range) follow-up was 9.4 (8.0–10.9) and 9.4 (7.8–11.0) years amongst men and women, respectively. DBP, diastolic blood pressure; SBP, systolic blood pressure.