Objectives: To investigate the association between blood pressure and mortality in very old people.
Design: Population-based cohort study.
Setting: County of Västerbotten, Sweden.
Participants: Half of all subjects aged 85 and all of those aged 90 and 95 and older (N=348) in one urban and five rural municipalities in the north of Sweden.
Measurements: Among others, supine blood pressure, Mini-Mental State Examination, Barthel Index of activities of daily living, Mini Nutritional Assessment, and body mass index. Information on diagnoses, medications, and 4-year mortality was collected. Associations between blood pressure and mortality were investigated using Cox regression analyses, controlling for a number of diagnoses and health factors.
Results: Baseline systolic blood pressure (SBP), diastolic blood pressure, and pulse pressure were all inversely associated with mortality within 4 years according to univariate analysis. SBP was the strongest predictor. In Cox regression analyses, low SBP (<or=120 mmHg) correlated with greater 4-year all-cause mortality alone and when controlling for health status. This connection persisted after exclusion of deaths within the first year. There was a tendency toward a U-shaped mortality curve for the adjusted model, with SBP of 164.2 mmHg (95% confidence interval=154.1-183.8 mmHg) being associated with the lowest mortality.
Conclusion: Lower SBP seems to be associated with greater mortality in people aged 85 and older, irrespective of health status. There are indications of a U-shaped correlation between SBP and mortality, and the optimal SBP for this age group could be above 140 mmHg.