National Trends in Antihypertensive Treatment Among Older Adults by Race and Presence of Comorbidity, 2008 to 2017

Abstract

Background: In 2014, hypertension guidelines for older adults endorsed increased use of fixed-dose combinations, prioritized thiazide diuretics and calcium channel blockers (CCBs) for Black patients, and no longer recommend beta-blockers as first-line therapy.

Objective: To evaluate older adults' antihypertensive use following guideline changes.

Design: Time series analysis.

Patients: Twenty percent national sample of Medicare Part D beneficiaries aged 66 years and older with hypertension.

Intervention: Eighth Joint National Committee (JNC8) guidelines MAIN MEASURES: Quarterly trends in prevalent and initial antihypertensive use were examined before (2008 to 2013) and after (2014 to 2017) JNC8. Analyses were conducted among all beneficiaries with hypertension, beneficiaries without chronic conditions that might influence antihypertensive selection (hypertension-only cohort), and among Black patients, given race-based guideline recommendations.

Key results: The number of beneficiaries with hypertension increased from 1,978,494 in 2008 to 2,809,680 in 2017, the proportions using antihypertensives increased from 80.3 to 81.2%, and the proportion using multiple classes and fixed-dose combinations declined (60.8 to 58.1% and 20.7 to 15.1%, respectively, all P<.01). Prior to JNC8, the use of angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, and CCBs was increasing. Use of CCBs as initial therapy increased more rapidly following JNC8 (relative change in quarterly trend 0.15% [95% CI, 0.13-0.18%), especially among Black beneficiaries (relative change 0.44% [95% CI, 0.21-0.68%]). Contrary to guidelines, the use of thiazides and combinations as initial therapy consistently decreased in the hypertension-only cohort (13.8 to 8.3% and 25.1 to 15.7% respectively). By 2017, 65.9% of Black patients in the hypertension-only cohort were initiated on recommended first-line or combination therapy compared to 80.3% of non-Black patients.

Conclusions: Many older adults, particularly Black patients, continue to be initiated on antihypertensive classes not recommended as first-line, indicating opportunities to improve the effectiveness and equity of hypertension care and potentially reduce antihypertensive regimen complexity.

Figure 1.

Trends in prevalent antihypertensive use, 2008 to 2017. Abbreviations: ACE/ARB, angiotensin-converting enzyme inhibitors and angiotensin receptor blockers; CCB, calcium channel blockers. Note: Hypertension-only cohorts include beneficiaries without a history of acute myocardial infarction, ischemic heart disease, atrial fibrillation, chronic kidney disease, diabetes, or heart failure. Proportions amount to greater than 100% due to concurrent use of multiple antihypertensives. A Change in trend comparing pre-JNC8 period to post-JNC8 period: P=.14 for thiazides, P<.001 for ACE/ARBs, beta blockers, CCBs, and other classes. B Change in trend comparing pre-JNC8 period to post-JNC8 period: P=.003 for ACE/ARBs, P=.001 for CCBs, P<.001 for beta blockers, thiazide diuretics, and other classes.

Figure 2.

Trends in prevalent antihypertensive use by race, 2008 to 2017. Abbreviations: ACE/ARB, angiotensin-converting enzyme inhibitors and angiotensin receptor blockers; CCB, calcium channel blockers. Note: Hypertension-only cohorts include beneficiaries without a history of acute myocardial infarction, ischemic heart disease, atrial fibrillation, chronic kidney disease, diabetes, or heart failure. Proportions amount to greater than 100% due to concurrent use of multiple antihypertensives. A Change in trend comparing pre-JNC8 period to post-JNC8 period: P<.001 for ACE/ARBs, P=.01 for beta blockers, P=.003 for CCBs, P=.29 for thiazide diuretics, P=.70 for other classes. B Change in trend comparing pre-JNC8 period to post-JNC8 period: P=.004 for CCBs, P<.001 for ACE/ARBs, beta blockers, thiazide diuretics, and other classes.

Figure 3.

Trends in choice of initial antihypertensive, 2008 to 2017. Abbreviations: ACE/ARB, angiotensin-converting enzyme inhibitors and angiotensin receptor blockers; CCB, calcium channel blockers. Note: Hypertension-only cohorts include beneficiaries without a history of acute myocardial infarction, ischemic heart disease, atrial fibrillation, chronic kidney disease, diabetes, or heart failure. Combinations include both fixed-dose combinations and separate antihypertensive classes initiated within the same seven day period. A Change in trend comparing pre-JNC8 period to post-JNC8 period: P=.28 for ACE/ARBs, P=.10 for beta blockers, P<.001 for CCBs, P<.001 for thiazide diuretics, P=.009 for other classes, P<.001 for combinations. B Change in trend comparing pre-JNC8 period to post-JNC8 period: P=.05 for ACE/ARBs, P=.008 for beta blockers, P<.001 for CCBs, P=.06 for thiazide diuretics, P=.17 for other classes, P<.001 for combinations.

Figure 4.

Trends in choice of initial antihypertensive by race, 2008 to 2017. Abbreviations: ACE/ARB, angiotensin-converting enzyme inhibitors and angiotensin receptor blockers; CCB, calcium channel blockers. Note: Hypertension-only cohorts include beneficiaries without a history of acute myocardial infarction, ischemic heart disease, atrial fibrillation, chronic kidney disease, diabetes, or heart failure. Combinations include both fixed-dose combinations and separate antihypertensive classes initiated within the same 7-day period. A Change in trend comparing pre-JNC8 period to post-JNC8 period: P=.06 for ACE/ARBs, P=.99 for beta blockers, P<.001 for CCBs, P=.75 for thiazide diuretics, P=.048 for other classes, P=.004 for combinations. B Change in trend comparing pre-JNC8 period to post-JNC8 period: P=.02 for ACE/ARBs, P=.006 for beta blockers, P<.001 for CCBs, P=.06 for thiazide diuretics, P=.42 for other classes, P<.001 for combinations.