Association between concurrent and remote blood pressure and disability in older adults

Abstract

The objective of this study was to investigate the association between blood pressure and disability in older adults. Stroke-free participants in the Charleston Heart Study (n=999, mean age=68.5+/- 0.2 years SE, 57% women, and 39% African Americans) were followed between 1960 and 1993. Functional measures including Nagi's Congruency in Medical and Self Assessment of Disability Scale, the Rosow-Breslaw Scale, and Katz' Activities of Daily Living Scale, in addition to systolic and diastolic blood pressures, were collected in 1984-1985, 1987-1990, and 1990-1993. Additional systolic and diastolic blood pressures from 1960 to 1963 were also available. We defined remote blood pressure change as the change from 1960 to 1984 to 1985 and concurrent blood pressure change as the change from 1984 to 1985 to the follow-up periods. Hypertension was defined as blood pressure >or=140/90 mm Hg or receiving antihypertensive agents, and it was considered uncontrolled if subjects were receiving antihypertensive agents and blood pressure was >or=140/90 mm Hg. Greater increases in remote and concurrent systolic blood pressure increases but not diastolic blood pressure were associated with greater declines in all 3 of the functional measures. Participants with hypertension were also at an increased risk for developing new disability (hazard ratio: 1.28, 95% CI: 1.04 to 1.59 for Nagi scale; hazard ratio: 1.28, 95% CI: 1.02 to 1.59 for Rosow-Breslaw Scale; and hazard ratio: 1.3, 95% CI: 1.01 to 1.69 for Katz scale). Participants with uncontrolled hypertension were at greatest risk of disability compared with normotensive subjects. In stroke-free older adults, increases in remote and concurrent systolic blood pressure increases are associated with greater functional decline. Older adults with uncontrolled hypertension are at a particularly increased risk for disability.

Figure 1

Change in functional outcome measures over the follow-up period in normotensives and hypertensive participants. Footnote: Estimates at each wave are the least square mean from the final Proc Mixed models with covariate adjustments: age, race, BMI, education, baseline measure, physical activity, and comorbidities. P-values are also from the Proc Mixed procedure testing the hypothesis Hº: rates of change in the outcome measures are different between the hypertensive and normotensives participants

Figure 2

Change in functional outcome measures over the follow-up period in normotensives, controlled hypertensive, and uncontrolled hypertensive participants Footnote: Estimates at each wave are the least square mean from the final Proc Mixed models with covariate adjustments (age, race, BMI, education, baseline measure, physical activity, and comorbidities. P-values are from the Proc Mixed procedure testing the hypotheses Hº: rate of change is different between normotensives and controlled hypertensives (p₁), rate of change is different between normotensives and uncontrolled hypertensives (p₂), and rate of change is different between controlled and uncontrolled hypertensives (p₃).