Talk to the Veterans Crisis Line now
U.S. flag
An official website of the United States government

VA Health Systems Research

Go to the VA ORD website
Go to the QUERI website
HSRD Conference Logo



2023 HSR&D/QUERI National Conference Abstract

Printable View

1092 — Outcomes of Antihypertensive Deintensification in Veteran Community Living Center Residents with Limited Life Expectancy and/or Advanced Dementia

Lead/Presenter: Ryan Hickson,  COIN - Pittsburgh/Philadelphia
All Authors: Hickson RP (Geriatric Research, Education, and Clinical Center, VA Pittsburgh Healthcare System, Pittsburgh, PA), Sileanu FE (Center for Health Equity Research and Promotion, VA Pittsburgh Healthcare System, Pittsburgh, PA) Zhao X(Center for Health Equity Research and Promotion, VA Pittsburgh Healthcare System, Pittsburgh, PA) Mor MK (Center for Health Equity Research and Promotion, VA Pittsburgh Healthcare System, Pittsburgh, PA) Niznik JD (Center for Health Equity Research and Promotion, VA Pittsburgh Healthcare System, Pittsburgh, PA) Aspinall SL (Center for Health Equity Research and Promotion, VA Pittsburgh Healthcare System, Pittsburgh, PA) Gellad WF (Center for Health Equity Research and Promotion, VA Pittsburgh Healthcare System, Pittsburgh, PA) Ersek M (Veterans Experience Center and the Center for Health Equity Research and Promotion, Corporal Michael J. Crescenz VA Medical Center, Philadelphia, PA) Schleiden LJ (Center for Health Equity Research and Promotion, VA Pittsburgh Healthcare System, Pittsburgh, PA) Springer SP (Center for Health Equity Research and Promotion, VA Pittsburgh Healthcare System, Pittsburgh, PA) Thorpe JM (Center for Health Equity Research and Promotion, VA Pittsburgh Healthcare System, Pittsburgh, PA) Hanlon JT (Geriatric Research, Education, and Clinical Center, VA Pittsburgh Healthcare System, Pittsburgh, PA) Thorpe CT (Center for Health Equity Research and Promotion, VA Pittsburgh Healthcare System, Pittsburgh, PA)

Objectives:
Objectives: Antihypertensive medications are frequently deintensified in older patients with limited life expectancy (LLE) and/or advanced dementia (AD) when therapy benefits are perceived to no longer outweigh risks. Evidence on health outcomes resulting from antihypertensive deintensification in this population is limited. Among Veterans with LLE/AD recently admitted to a VA nursing home (Community Living Center, or CLC), we estimated the impact of antihypertensive deintensification on all-cause emergency department visits (ED), hospitalizations, and death.

Methods:
We used fiscal year 2009-2015 VA and Medicare data to conduct a national retrospective cohort study of CLC residents with LLE/AD who were potentially overtreated for hypertension, defined as receiving at least 1 antihypertensive medication and having systolic blood pressure < 120mmHg. Deintensification required a 7-day sustained dose decrease or discontinuation of at least 1 antihypertensive. Entropy balance weighting was used to address confounding when comparing outcomes of antihypertensive regimen deintensification vs. continuation. Weighted Aalen-Johansen cumulative incidence curves (treating ED/hospitalization and death as separate outcomes) estimated 30- and 60-day relative risks (RRs) of antihypertensive deintensification vs. continuation, with nonparametric bootstrap 95% confidence intervals (95%CI). Underlying causes of death and ED/hospitalization were explored descriptively. All analyses were stratified by receipt of hospice care.

Results:
Among 5,333 non-hospice residents who were potentially overtreated, 37.1% deintensified antihypertensives after CLC admission. Comparing deintensification vs. continuing antihypertensives after entropy balance weighting, the 30-day RR of ED/hospitalization was 1.03 (95%CI:0.90-1.19); the 60-day risk was consistent (RR = 1.08; 95%CI:0.96-1.20). The 30-day RR of death from antihypertensive deintensification vs. continuation was 1.01 (95%CI:0.76-1.45); the 60-day RR was 1.17 (95%CI:0.93-1.58). Among 2,968 Veterans receiving hospice care, 46.0% deintensified antihypertensives after CLC admission. After entropy balance weighting, the 30-day RR for ED/hospitalization from antihypertensive deintensification vs. continuation was 0.73 (95%CI:0.49-1.09); the 60-day risk was consistent (RR = 0.80; 95%CI:0.56-1.16). The 30-day RR for death was 1.14 (95%CI:1.05-1.25); the 60-day risk was consistent (RR = 1.14; 95%CI:1.07-1.22). Veterans receiving hospice care who deintensified vs. continued antihypertensives were more likely to die from cancer (38.1% vs. 30.3%, absolute standardized difference: 16.6%).

Implications:
We found no evidence of increased harm from antihypertensive deintensification among non-hospice CLC residents with LLE/AD who were potentially overtreated. Among CLC residents receiving hospice care, unmeasured confounding by survival prognosis may explain the association of antihypertensive deintensification with death; cancer deaths were more common in CLC residents who deintensified vs. continued their antihypertensive regimen which is unlikely to be caused by antihypertensive deintensification.

Impacts:
Among CLC residents with LLE/AD who are not receiving hospice care, our findings support recommendations to deintensify antihypertensive therapy to align with palliative goals of care in Veterans who are potentially overtreated. The outcome of death is expected in end-of-life populations but often must be considered as a competing risk for other clinically meaningful outcomes, like avoidance of burdensome transfers to acute care settings. Additional study design considerations, such as requiring more time to define exposures, may be needed to appropriately estimate the impact of treatment decisions with observational data in end-of-life populations when a valid measure of short-term survival prognosis is unavailable to address confounding.