Survival curves based on Cox proportional hazard regression models

are shown for systolic BP in Figure 2 and diastolic BP in eFigure 1. In initial age- and sex-adjusted analysis of the total sample (model 1), compared with systolic BP ≤ 125 mm Hg, mortality risk decreased with increasing BP category (126–139 mm Hg: HR 0.70, 95% confidence interval [CI] 0.53–0.93; 140–149 mm Hg: HR 0.63, 95% CI 0.48–0.83; 150–164 mm Hg: HR 0.59, 95% CI 0.45–0.76; ≥165 mm Hg: HR 0.50, 95% CI 0.38–0.66; Table 3, Figure 2A). BMN 673 ic50 None of these associations was significant in the fully adjusted model for the total sample (model 2). For diastolic BP, mortality risk was significantly increased in the quartile of BP lower than 70 mm Hg in model 1 (HR 1.42, 95% CI 1.11–1.81) and in the quartile of 70 to 74 mm Hg in models 1 (HR 1.32, 95% CI 1.03–1.69) and 2 (HR 1.37, 95% CI 1.03–1.83), compared with the quartile of 75 to 80 mm Hg (eTable 1, eFigure 1A). The association of BP with mortality differed among gait speed subcohorts. In the slower-walking subcohort, patterns of association were similar to those

of the total sample (Table 3, Figure 2B, eTable 1, eFigure 1B). In age- and sex-adjusted analysis of the faster-walking subcohort, mortality risk was more than twice higher in participants with systolic BP of 125 mm Hg or lower than in those with systolic BP of 126 to 139 mm Hg (HR 2.38, 95% CI 1.05–5.41). This association did not reach statistical significance in the fully adjusted Farnesyltransferase analysis (model 2); instead, mortality risk http://www.selleckchem.com/products/abt-199.html was more than twice higher in participants with systolic BP of 165 mm Hg or higher (HR 2.13, 95% CI 1.01–4.49) and 140 to 149 mm Hg (HR 2.25, 95% CI 1.03–4.94) than in those with systolic BP of 126 to 139 mm

Hg (Table 3, Figure 2C). For diastolic BP, mortality was significantly higher in the highest quartile (>80 mm Hg) in models 1 (HR 1.65, 95% CI 1.01–2.69) and 2 (HR 1.76, 95% CI 1.07–2.90) compared with the quartile of 75 to 80 mm Hg in the faster-walking subcohort (eTable 1, eFigure 1C). In the age- and sex-adjusted analysis, interaction effects between gait speed subcohort and BP in the association with mortality were significant for systolic BP (P = .031), but not for diastolic BP (P = .283). Interaction effects were not significant for systolic BP (P = .327) or diastolic BP (P = .272) in the fully adjusted model. Repeated analyses with the exclusion of data from participants who died in the first year of study inclusion produced essentially the same results (data not shown). In this study of a representative sample of very old individuals, low systolic and diastolic BP were significantly associated with increased mortality risk in initial age- and sex-adjusted analyses, but not in analyses adjusted for all covariates, including previous disease.