Prevalence and Prognosis of Blood Pressure Medication Deintensification among Older VA Nursing Home Residents

Northern California Institute/Res/Edu, San Francisco CA

Investigators

Linked publications, trials & patents

Abstract

ABSTRACT Persons age >65 years are the fastest growing segment of the U.S. population, and nearly half will enter a nursing home before they die. The vast majority of nursing home residents have high blood pressure (BP). Evidence to inform management of high BP in this population is limited because persons living in nursing homes have been excluded from randomized trials evaluating the benefits and risks of lowering BP. Clinicians treating high BP in nursing home residents must extrapolate data from trials including elders with a smaller burden of concomitant chronic conditions, and better functional status and cognitive function. Guidelines suggest caution when initiating or intensifying antihypertensive medications in vulnerable elders and some advocate higher BP goals for persons with limited functional status. However, once an elder is treated to low BP, there is no guidance on whether or not treatment should be deintensified, defined as drug discontinuation or dosage decrease. Treatment deintensification has the potential to reduce harms from excessive BP lowering in vulnerable elders. The primary goals of this research are to (1) conduct a detailed and rigorous epidemiologic study of current high BP management among Veterans residing in VA nursing homes; and (2) use state-of-the art causal inference methods to evaluate the prognosis of deintensification in this population. Our overall hypothesis is that very low BP is harmful in this population, and that antihypertensive medication deintensification is underutilized, even among persons with very low BP, dementia, exceptional multimorbidity, or poor functional status. We further hypothesize that deintensification will be associated with preservation of functional and cognitive status without increasing cardiovascular risk. We plan to use national VA data, linked to Centers for Medicare & Medicaid Services (CMS) data because, unlike prior studies, VA data include risk factor measurements (e.g. BP), laboratory measures (e.g. renal function), and detailed medication data. This research has the following aims: (1) to evaluate the association of BP level and number of antihypertensive medications with patient centered outcomes and clinical outcomes;(2) to assess rates of deintensification of antihypertensive medications overall, and among those potentially overtreated, and to ascertain patient characteristics associated with deintensification; and (3) to determine the association of deintensification vs. continued treatment with patient centered outcomes and clinical outcomes, using novel causal inference methods to account for confounding by indication.