Time in Therapeutic Range, as a Determinant of All-Cause Mortality in Patients With Hypertension

Abstract

Background: Accumulating evidence indicates that reducing systolic blood pressure (BP) to <140 mm Hg improves health outcomes; however, an optimal level has not yet been determined. Many population studies or post hoc analyses suggest a target systolic BP between 120 and 140 mm Hg with increased risk above and below that range. We tested the hypothesis that consistent control of systolic BP between 120 and 140 mm Hg-time in therapeutic range-is a strong determinant of all-cause mortality among US veterans.

Methods and results: A total of 689 051 individuals from 15 Veterans Administration Medical Centers were followed over a 10-year period. Participants were classified as hypertensive, intermediate hypertensive, and normotensive according to the number of elevated BP recordings (>3, 1 or 2, and none, respectively). Time within, above, or below therapeutic range (120-140 mm Hg) was considered in quartiles and related to all-cause mortality. The study population consisted of 54% hypertensive, 19.9% intermediate, and 26.1% normotensive participants; the corresponding mortality rates for the 3 groups were 11.5%, 8%, and 1.9%, respectively (P<0.0001 for all comparisons). Mortality rates for hypertensive participants with BP consistently within, above, or below the therapeutic range were 6.5%, 21.9%, and 33.1%, respectively (P<0.0001 for all comparisons). Mortality rates in hypertensive participants increased from 6.5% in the most consistently controlled quartile (>75%) to 8.9%, 15.6%, and 23.5% towards the less consistently controlled quartiles (50-75%, 25-50%, and <25%, respectively; P<0.0001 for all comparisons).

Conclusions: An inverse and gradual association between time in therapeutic range and all-cause mortality was observed in this large veteran cohort. Consistency of BP control over time is a strong determinant of all-cause mortality, and consistency of BP control should be monitored in everyday clinical practice.

Keywords: hypertension; mortality; variation.

Figure 1

A, Death rates by last systolic BP in patients with established hypertension (>3 elevated BPs). B, Death rates by last systolic BP in patients with only 2 elevated systolic BPs (MID‐hypertensive group). C, Death rates by last systolic BP in nonhypertensives. D, Mortality rates by last systolic BP between 120 and 140 mmHg for Normal, Mid HTN, Hypertension, by 10 mmHg increments. E, Mortality rates by last systolic BP between 120 and 140 mmHg for Normal, Mid HTN, Hypertension, by 20mmHg increments. BP indicates blood pressure; HTN, hypertension; SBP, systolic blood pressure.

Figure 2

A, Mortality rates among patients with established hypertension based on percent of readings in therapeutic range. B, Mortality rates among patients with MID‐hypertension based on percent of readings in therapeutic range. C, Mortality rates among never‐hypertensive patients based on percent of readings in therapeutic range. BP indicates blood pressure.

Figure 3

Cox regression models: mortality rates for patients with systolic BP measurements in therapeutic range 0% to 25%, 25% to 50%, 50% to 75%, or >75% of BP measurements. Follow‐up to 120 mo. Corrected for location, age, sex, and race. BP indicates blood pressure; SBP, systolic blood pressure.