Trends in Use of High-Intensity Statin Therapy After Myocardial Infarction, 2011 to 2014

JOURNAL OF THE AMERICAN COLLEGE OF CARDIOLOGY

VOL. 69, NO. 22, 2017

ª 2017 BY THE AMERICAN COLLEGE OF CARDIOLOGY FOUNDATION

ISSN 0735-1097/$36.00

PUBLISHED BY ELSEVIER

http://dx.doi.org/10.1016/j.jacc.2017.03.585

Trends in Use of High-Intensity Statin Therapy After Myocardial Infarction, 2011 to 2014 Robert S. Rosenson, MD,a Michael E. Farkouh, MD,b Matthew Mefford, MS,c Vera Bittner, MD,d Todd M. Brown, MD, MSPH,d Ben Taylor, PHD,e Keri L. Monda, PHD,e Hong Zhao, MSPH,c Yuling Dai, MSPH,c Paul Muntner, PHDc

ABSTRACT BACKGROUND Data prior to 2011 suggest that a low percentage of patients hospitalized for acute coronary syndromes filled high-intensity statin prescriptions upon discharge. Black-box warnings, generic availability of atorvastatin, and updated guidelines may have resulted in a change in high-intensity statin use. OBJECTIVES The aim of this study was to examine trends and predictors of high-intensity statin use following hospital discharge for myocardial infarction (MI) between 2011 and 2014. METHODS Secular trends in high-intensity statin use following hospital discharge for MI were analyzed among patients 19 to 64 years of age with commercial health insurance in the MarketScan database (n ¼ 42,893) and 66 to 75 years of age with U.S. government health insurance through Medicare (n ¼ 75,096). Patients filling statin prescriptions within 30 days of discharge were included. High-intensity statins included atorvastatin 40 or 80 mg and rosuvastatin 20 or 40 mg. RESULTS The percentage of beneficiaries whose first statin prescriptions filled following hospital discharge for MI were for high-intensity doses increased from 33.5% in January through March 2011 to 71.7% in October through November 2014 in MarketScan and from 24.8% to 57.5% in Medicare. Increases in high-intensity statin use following hospital discharge occurred over this period among patients initiating treatment (30.6% to 72.0% in MarketScan and 21.1% to 58.8% in Medicare) and those taking low- or moderate-intensity statins prior to hospitalization (from 27.8% to 62.3% in MarketScan and from 12.6% to 45.1% in Medicare). In 2014, factors associated with filling high-intensity statin prescriptions included male sex, filling beta-blocker and antiplatelet agent prescriptions, and attending cardiac rehabilitation within 30 days following discharge. CONCLUSIONS The use of high-intensity statins following hospitalization for MI increased progressively from 2011 through 2014. (J Am Coll Cardiol 2017;69:2696–706) © 2017 by the American College of Cardiology Foundation.

S

everal randomized controlled trials have shown

(CVD) events among patients hospitalized for acute

high-intensity statin therapy to be more effica-

coronary syndrome (ACS) (1–3). Analyses of registries

cious than lower intensity therapy for prevent-

and insurance claims databases have documented

ing recurrent atherosclerotic cardiovascular disease

that between 20% and 40% of patients with ACS fill

From aMount Sinai Heart, Icahn School of Medicine at Mount Sinai, New York, New York; bPeter Munk Cardiac Centre and Heart and Stroke Richard Lewar Centre, University of Toronto, Toronto, Ontario, Canada; cDepartment of Epidemiology University of Alabama at Birmingham, Birmingham, Alabama; dDepartment of Medicine, Division of Cardiovascular Disease, University of Alabama at Birmingham, Birmingham, Alabama; and the eCenter for Observational Research, Amgen, Thousand Oaks, California. Listen to this manuscript’s

The present study was funded by an industry-academic collaboration between Amgen, the University of Alabama at Birmingham,

audio summary by

and the Icahn School of Medicine at Mount Sinai. Dr. Rosenson has received grant support from Akcea, Amgen, AstraZeneca, Eli

JACC Editor-in-Chief

Lilly, Esperion, The Medicines Company, and Sanofi; serves on advisory boards for Amgen, Eli Lilly, Regeneron, and Sanofi; has

Dr. Valentin Fuster.

received honoraria from Akcea and Kowa; and has received royalties from UpToDate. Dr. Farkouh has received grant support from Amgen. Dr. Bittner has received grant support from Amgen, AstraZeneca, DalCor, Sanofi-Regeneron, Pfizer, and Bayer Healthcare; and has served on advisory boards for Amgen and Eli Lilly. Dr. Brown has received grant support from Amgen and AstraZeneca. Dr. Taylor is employed by Amgen. Dr. Monda is employed by and holds stock in Amgen. Dr. Muntner has received grant support from Amgen; and honoraria from Amgen. Freny Vaghaiwalla Mody, MD, served as Guest Editor for this paper. Manuscript received March 15, 2017; accepted March 31, 2017.

Rosenson et al.

JACC VOL. 69, NO. 22, 2017 JUNE 6, 2017:2696–706

2697

Trends in the Use of High-Intensity Statin Therapy

prescriptions for high-intensity statins following hos-

obtained through the Truven Health Market-

ABBREVIATIONS

pital discharge (4,5). However, these studies were

Scan Research Database. Medicare is a gov-

AND ACRONYMS

conducted among patients having coronary heart dis-

ernment

ease (CHD) events through 2011, with few more

insurance for U.S. adults $65 years of age and

contemporary data being published.

younger adults with end-stage renal disease

SEE PAGE 2707

program

that

provides

health

ACC = American College of Cardiology

ACS = acute coronary

or who are disabled. Administrative claims

syndrome

for Medicare beneficiaries were obtained

AHA = American Heart

Several events have occurred over the past several

from the Chronic Conditions Warehouse,

Association

years that may have changed the use of high-intensity

which was created by the Centers for Medi-

CHD = coronary heart disease

statins among patients with ACS. In 2011, the

care and Medicaid Services to provide data

CVD = cardiovascular disease

American Heart Association (AHA)/American College

for research purposes. The Institutional Re-

MI = myocardial infarction

of Cardiology Foundation

view Board at the University of Alabama at

secondary prevention

guidelines recommended “adequate” doses of statin

Birmingham approved this analysis of deidentified

therapy necessary to achieve specific low-density

data.

lipoprotein cholesterol thresholds (6). Although no

We studied MarketScan and Medicare beneficiaries

specific statin intensity was recommended for pa-

who were hospitalized with overnight stays for MI

tients with ACS, this guideline stated that it was

between January 1, 2011, and November 30, 2014

reasonable (Class IIa, Level of Evidence: C) to treat

(Online Figures 1 and 2). We restricted the analyses to

patients who were at very high risk, including those

patients whose hospital stays for their index MIs

with ACS, to an low-density lipoprotein cholesterol

were #30 days, who were alive 30 days following

level <70 mg/dl. In 2013, the American College of

hospital discharge, who had continuous MarketScan

Cardiology (ACC)/AHA guideline on the treatment of

or Medicare fee-for-service insurance, and who were

blood cholesterol to reduce atherosclerotic CVD risk

living in the United States from 365 days prior to

in adults was published and recommended the use of

hospital admission for their MIs through 30 days

high-intensity statin therapy for patients with estab-

following

lished CVD (1). Other considerations that may have

coverage includes Parts A (inpatient), B (outpatient), D

discharge. (prescription).

Medicare

fee-for-service

influenced prescribing patterns beginning in 2011

and

include a black-box warning against new pre-

Medicare Advantage plans (Medicare Part C) were

Beneficiaries

enrolled

in

scriptions for simvastatin 80 mg and the generic

excluded, because complete claims are not available

availability of atorvastatin.

for these persons. We excluded MarketScan and

We evaluated secular trends in high-intensity

Medicare beneficiaries with stays at skilled nursing

statin use following hospitalization for myocardial

facilities or hospice facilities within 30 days following

infarction (MI) from 2011 through 2014 in 2 large

their index MIs or with outpatient statin fills during

samples of U.S. adults. Also, we identified patient

their MI hospitalizations. To avoid overlap between

characteristics for those who initiated statins with a

the 2 samples, we restricted the MarketScan analysis

high- versus a low- or moderate-intensity dose

to beneficiaries <65 years of age and Medicare anal-

following their MIs and those who were titrated

yses to beneficiaries 66 to 75 years of age at the time

from low or moderate statin intensity to a high-

of hospital admission for MI. An age $66 years was

intensity statin after their MIs in 2014. The results

used, rather than $65 years, to allow a 1-year look-

of this study provide contemporary data on the use

back period to identify characteristics of Medicare

of high-intensity statins following publication of

beneficiaries. We analyzed Medicare beneficiaries

the most recent ACC/AHA cholesterol guidelines,

>75 years of age separately because there are only

characterize patients who do versus do not receive

limited data supporting high-intensity statins for this

high-intensity statins following hospital discharge

group. We restricted these analyses to beneficiaries

for MI, and thereby identify those with an unmet

who

treatment need.

following hospital discharge for MI. The first MI each

filled

statin

prescriptions

within

30

days

beneficiary experienced that met these criteria was

METHODS

included and is referred to as the index MI.

We conducted a retrospective cohort study using

STATIN USE. Statin use was identified by pharmacy

administrative claims from MarketScan and Medi-

prescription fills in MarketScan claims and Medicare

care. The MarketScan database contains health care

Part D pharmacy claims in combination with national

claims for persons with commercial, Medicare sup-

drug codes. Statins included atorvastatin, fluvastatin,

plemental, and Medicaid health insurance and was

lovastatin, pitavastatin, pravastatin, rosuvastatin,

2698

Rosenson et al.

JACC VOL. 69, NO. 22, 2017 JUNE 6, 2017:2696–706

Trends in the Use of High-Intensity Statin Therapy

and simvastatin. High-intensity statins use prior to

and November to allow a 30-day follow-up period

the index MI included atorvastatin 40 or 80 mg,

to identify statin prescription fills post–hospital

rosuvastatin 20 or 40 mg, and simvastatin 80 mg (1).

discharge). For each quarter, we assessed the first

Because simvastatin 80 mg is no longer recom-

type of statin prescription and dose filled following

mended for patients with ACS, MarketScan and

hospital discharge for each beneficiary’s index MI. As

Medicare beneficiaries filling prescriptions for sim-

mentioned earlier, we excluded from the remaining

vastatin 80 mg were excluded for analyses of high-

analyses beneficiaries (n ¼ 754 in MarketScan and

intensity statin use following MI (see statistical

n ¼ 1,476 in Medicare) whose first statin prescription

methods). Beneficiaries with any statin fills in the

fill following hospital discharge was for simvastatin

365 days prior to their index MI were categorized as

80 mg (1). We calculated the percentage of Market-

prevalent statin users. The statin intensity among

Scan

prevalent users was based on the fill most proximate

following their index MIs were for high-intensity

to, yet preceding, each beneficiary’s index MI.

statins. This was done for the overall population,

BENEFICIARY CHARACTERISTICS. Age and sex were

among statin initiators, among low- and moderate-

identified from the MarketScan Commercial Claims

intensity prevalent users, and among high-intensity

and Encounters database and the Medicare benefi-

prevalent statin users.

ciary summary file. Each beneficiary’s age was calcu-

beneficiaries

whose

first

prescription

fills

Characteristics of MarketScan beneficiaries filling

lated on the date of his or her index MI. Information on

and not filling high-intensity statin prescriptions

race/ethnicity is available for Medicare beneficiaries

following hospital discharge for MI in 2014 were

through the beneficiary summary file, but information

calculated among statin initiators and among preva-

on race is not available for MarketScan beneficiaries.

lent low- and moderate-intensity and among high-

In Medicare, receipt of a low-income subsidy under

intensity statin users separately. This analysis was

Medicare Part D and Medicare/Medicaid dual eligi-

restricted to 2014 to provide data for the time period

bility from the Medicare beneficiary enrollment file

after publication of the ACC/AHA cholesterol man-

were used as markers of low socioeconomic status.

agement guideline. We used Poisson regression with

Diabetes, CHD, stroke, heart failure, peripheral artery

sandwich estimators to calculate the relative risks

disease, chronic kidney disease, depression, Charlson

for a high-intensity statin as the first statin prescrip-

comorbidity index, all-cause hospitalizations, cardi-

tion fill within 30 days after discharge. Relative risks

ologist care, use of nonstatin lipid-lowering therapy,

were calculated in an unadjusted model, after demo-

and the total number of different medication pre-

graphic adjustment (age and sex in MarketScan

scriptions filled were identified using claims in the 365

and age, sex, and race/ethnicity in Medicare) and in

days prior to hospital admission for MI and previously

a model that included age, sex, race (Medicare only),

published algorithms (Online Table 1) (7–13). The

low income or dual subsidy (Medicare only), history

presence of dementia was determined for Medicare

of diabetes, CHD, stroke, heart failure, peripheral

beneficiaries but was not included in the analysis of

artery disease, chronic kidney disease, dementia

MarketScan beneficiaries given its low prevalence

(Medicare only), Charlson comorbidity index, depres-

(<1%) in this younger population. We identified the

sion, hospitalization in the year prior to MI, cardiolo-

use of cardiac rehabilitation and prescription fills for

gist care, nonstatin lipid-lowering therapy use, total

beta-blockers, antiplatelet agents, and nonstatin

number of medications taken, 30-day post-discharge

lipid-lowering therapies within 30 days after hospital

cardiac rehabilitation, or 30-day post-discharge fills

discharge for the index MI.

for beta-blockers, antiplatelet agents, and nonstatin lipid-lowering therapy, simultaneously.

STATISTICAL ANALYSIS. All analyses were conduct-

We calculated the percentage of beneficiaries

ed for MarketScan beneficiaries <65 years of age and

who switched from low- or moderate-intensity to high-

Medicare beneficiaries 66 to 75 years of age sepa-

intensity statins within 182 days following discharge.

rately. Below, we describe the analyses for Market-

To allow 182 days of follow-up, this analysis was

Scan beneficiaries. Identical analyses were conducted

restricted to beneficiaries with index MIs in the first

for Medicare beneficiaries. Characteristics of benefi-

quarter of 2011 through the second quarter of 2014;

ciaries included in this analysis were calculated

those who died or lost insurance coverage within 182

overall and for each calendar year. The study period

days of hospital discharge for their index MI were

from 2011 through 2014 was divided into calendar

excluded. This analysis included 17,574 MarketScan

quarters (January to March, April to June, July to

beneficiaries and 37,213 Medicare beneficiaries whose

September, and October to December, except 2014,

first statin prescription fills following hospital discharge

for which the fourth quarter included only October

were for low- or moderate-intensity statin doses.

Rosenson et al.

JACC VOL. 69, NO. 22, 2017 JUNE 6, 2017:2696–706

2699

Trends in the Use of High-Intensity Statin Therapy

F I G U R E 1 Type of Statin Filled After Hospital Discharge for Myocardial Infarction Among MarketScan Beneficiaries <65 Years of Age and Medicare Beneficiaries

66 to 75 Years of Age

MarketScan Beneficiaries

Medicare Beneficiaries

100%

90%

90%

80%

80%

70%

70%

60%

60%

Percent (%)

Percent (%)

100%

50% 40%

50% 40%

30%

30%

20%

20%

10%

10%

0%

0% Q1

Q2 Q3 Q4

Q1

Q2 Q3 Q4

2011

Q1

2012

Q2 Q3 Q4

Q1

2013

Atorvastatin

Simvastatin

Q2 Q3 Q4 2014 Rosuvastatin

Q1

Q2 Q3 Q4 Q1

Q2 Q3 Q4 Q1

2011

2012

Pravastatin

Lovastatin

Q2 Q3 Q4 Q1 2013

Q2 Q3 Q4 2014

Other

Among MarketScan beneficiaries (left), an increase in atorvastatin prescriptions and a decrease in simvastatin prescriptions was observed between quarter 1 of 2011 and quarter 4 of 2014. Other statin use remained steady. Similar patterns were observed among Medicare beneficiaries (right). MI ¼ myocardial infarction; Q1 ¼ calendar quarter 1; Q2 ¼ calendar quarter 2; Q3 ¼ calendar quarter 3; Q4 ¼ calendar quarter 4.

In a final analysis, we assessed the type of statin

The percentage of beneficiaries whose first statin

prescription and dose filled following hospital discharge

prescription fills following hospital discharge for MI

for MI among Medicare beneficiaries >75 years of age for

were for high-intensity doses increased from 33.5% to

each calendar quarter in 2011 through 2014. We also

71.7% among MarketScan beneficiaries and from

calculated the percentage of Medicare beneficiaries >75

24.8% to 57.5% among Medicare beneficiaries 66 to

years of age whose first statin fill following their index

75 years of age (Central Illustration, Online Table 7).

MI was for a high-intensity dose. All data management

An increase in the percentage of beneficiaries filling

and statistical analyses were conducted using SAS

high-intensity statin prescriptions following hospital

version 9.4 (SAS Institute, Cary, North Carolina).

discharge for MI occurred for MarketScan and Medi-

RESULTS

72.0% and from 21.1% to 58.8%, respectively) and

care beneficiaries initiating statins (from 30.6% to those who were taking low- or moderate-intensity TRENDS IN STATIN PRESCRIPTION FILLS. The cur-

statins (from 27.8% to 62.3% and from 12.6% to

rent

MarketScan

45.1%, respectively) prior to their index MIs. Among

beneficiaries <65 years of age and 75,096 Medicare

prevalent high-intensity statin users prior to their MI

analysis

included

42,893

beneficiaries 66 to 75 years of age who filled statin

hospitalizations, the percentage filling high-intensity

prescriptions within 30 days following hospital

statin prescriptions following hospital discharge was

discharge for MI. A higher proportion of MarketScan

86.1% in the first quarter of 2011 and 90.6% in the

beneficiaries were male, and a higher proportion of

fourth quarter of 2014 for MarketScan beneficiaries

Medicare beneficiaries had comorbidities (Online

and 85.0% and 90.3%, respectively, among Medicare

Table 2). Beneficiary characteristics in MarketScan

beneficiaries 66 to 75 years of age.

and Medicare are presented by calendar year in

FACTORS

Online Tables 3 and 4. Between 2011 and 2014, the

STATIN FILLS IN 2014. Demographic characteristics,

ASSOCIATED

WITH

HIGH-INTENSITY

percentage of MarketScan and Medicare beneficiaries

the prevalence of comorbidities, and health care use

whose first statin prescription fills following hospital

are presented by intensity of statin prescriptions

discharge

rosuvastatin

filled following hospital discharge for MI in 2014

decreased, whereas prescription fills for atorvastatin

among MarketScan and Medicare beneficiaries, 66 to

increased (Figure 1, Online Tables 5 and 6).

75 years of age, initiating treatment and for those

were

for

simvastatin

or

Rosenson et al.

2700

JACC VOL. 69, NO. 22, 2017 JUNE 6, 2017:2696–706

Trends in the Use of High-Intensity Statin Therapy

C E NT R AL IL L U STR AT IO N Trends in the Use of High-Intensity Statin Therapy MarketScan Beneficiaries

Medicare Beneficiaries 100%

100% 90%

90%

80%

80% 70%

Percent (%)

Percent (%)

70% 60% 50% 40%

60% 50% 40%

30%

30%

20%

20%

10%

10% 0%

0%

Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4

Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 2011

2012

2013

Overall

2011

2014

Statin Initiators

Low/Moderate Intensity Statin Users Prior to MI

2012

2013

2014

High Intensity Statin Users Prior to MI

Rosenson, R.S. et al. J Am Coll Cardiol. 2017;69(22):2696–706.

High-intensity statin prescription fills after hospital discharge for myocardial infarction (MI) among MarketScan beneficiaries <65 years of age (left) and Medicare beneficiaries 66 to 75 years of age (right). Among MarketScan beneficiaries, an increase in high-intensity statin use was observed for beneficiaries initiating statin therapy and prevalent low- or moderate-intensity statin users prior to MI hospitalization. High-intensity statin use following MI hospitalization remained steady among prevalent high-intensity users. Similar patterns were observed among Medicare beneficiaries. Q1 ¼ calendar quarter 1; Q2 ¼ calendar quarter 2; Q3 ¼ calendar quarter 3; Q4 ¼ calendar quarter 4.

taking low- or moderate-intensity and high-intensity

discharge (Online Table 12). Among Medicare benefi-

statins prior to their MIs in Tables 1 and 2, respec-

ciaries taking high-intensity statins prior to their

tively. After multivariate adjustment and among both

MIs, men were more likely than women to fill

MarketScan and Medicare beneficiaries, men, those

high-intensity statin prescriptions post-discharge

filling

(Online Table 13).

beta-blocker

and

antiplatelet

agent

pre-

scriptions within 30 days following discharge, and those attending cardiac rehabilitation after their MIs were more likely to initiate statins with a high- versus low- or moderate-intensity dose or titrate from a lowor

moderate-intensity

to

a

high-intensity

dose

(Figure 2, Online Tables 8 to 11). Beneficiaries taking more medications prior to their index MIs were less likely to initiate statin therapy with high-intensity doses. Among Medicare beneficiaries, African Americans were more likely than whites to initiate treatment with high-intensity statins following their MIs. MarketScan and Medicare beneficiaries with histories of CHD, cardiologist care in the year prior to their index

TITRATION TO A HIGH-INTENSITY STATIN WITHIN 182 DAYS OF HOSPITAL DISCHARGE. Among those

whose first statin prescription fills post-discharge were for low- or moderate-intensity doses, the percentage who up-titrated to high-intensity statins within 182 days of hospital discharge increased between the first quarter of 2011 and the second quarter of 2014 from 6.1% to 12.9% among MarketScan beneficiaries (Online Table 14, left column) and from 3.7% to 9.2% among Medicare beneficiaries (Online Table 14, right column). HIGH-INTENSITY STATIN USE AMONG MEDICARE

MIs, and filling prescriptions for nonstatin lipid-

BENEFICIARIES >75 YEARS OF AGE. Among Medi-

lowering therapy within 30 days of discharge were

care beneficiaries >75 years of age, the percentage

less likely to titrate from a low- or moderate-intensity

whose first statin prescription fills following hospital

to a high-intensity dose. Among MarketScan benefi-

discharge were for simvastatin decreased from 55.1%

ciaries taking high-intensity statins prior to their

in the first quarter of 2011 to 14.6% in the fourth

MIs, no factors studied were associated with filling

quarter of 2014, while atorvastatin prescriptions

high-intensity statin prescriptions following hospital

increased from 22.8% to 66.3% (Online Figure 3, left).

Rosenson et al.

JACC VOL. 69, NO. 22, 2017 JUNE 6, 2017:2696–706

Trends in the Use of High-Intensity Statin Therapy

T A B L E 1 Characteristics of MarketScan Beneficiaries <65 Years of Age Filling and Not Filling High-Intensity Statin Prescriptions Following

Hospital Discharge for Myocardial Infarction in 2014

Statin Initiators (n ¼ 6,303)

Prevalent Low- or ModerateIntensity Statin Users (n ¼ 2,234)

Prevalent High-Intensity Statin Users (n ¼ 870)

High-Intensity Statin† Fill

High-Intensity Statin Fill

High-Intensity Statin Fill

No (n ¼ 2,041)

Yes (n ¼ 4,262)

No (n ¼ 987)

Yes (n ¼ 1,247)

No (n ¼ 78)

Yes (n ¼ 792)

Age, yrs <40

98 (4.8)

171 (4.0)

8 (0.8)

19 (1.5)

1 (1.3)

20 (2.5)

40–49

406 (19.9)

934 (21.9)

118 (12.0)

172 (13.8)

14 (17.9)

104 (13.1)

50–59

949 (46.5)

2,052 (48.1)

455 (46.1)

618 (49.6)

34 (43.6)

378 (47.7)

60–64

588 (28.8)

1,105 (25.9)

406 (41.1)

438 (35.1)

29 (37.2)

290 (36.6)

Male

1,449 (71.0)

3,278 (76.9)

703 (71.2)

972 (77.9)

55 (70.5)

632 (79.8)

History of diabetes

295 (14.5)

534 (12.5)

375 (38.0)

448 (35.9)

33 (42.3)

355 (44.8)

History of CHD

217 (10.6)

400 (9.4)

295 (29.9)

218 (17.5)

33 (42.3)

315 (39.8)

History of stroke

15 (0.7)

23 (0.5)

43 (4.4)

26 (2.1)

4 (5.1)

23 (2.9)

History of heart failure

23 (1.1)

40 (0.9)

47 (4.8)

35 (2.8)

7 (9.0)

47 (5.9)

History of PAD

10 (0.5)

26 (0.6)

28 (2.8)

26 (2.1)

4 (5.1)

36 (4.5)

History of CKD

97 (4.8)

156 (3.7)

136 (13.8)

94 (7.5)

15 (19.2)

118 (14.9)

1,344 (65.9)

3,006 (70.5)

363 (36.8)

589 (47.2)

29 (37.2)

271 (34.2)

587 (28.8)

1,102 (25.9)

471 (47.7)

531 (42.6)

34 (43.6)

383 (48.4) 138 (17.4)

Charlson comorbidity index 0 1–3 $4

110 (5.4)

154 (3.6)

153 (15.5)

127 (10.2)

15 (19.2)

Depression

130 (6.4)

254 (6.0)

104 (10.5)

101 (8.1)

6 (7.7)

65 (8.2)

Hospitalization in year prior to MI

146 (7.2)

227 (5.3)

179 (18.1)

126 (10.1)

17 (21.8)

137 (17.3)

447 (21.9)

765 (17.9)

459 (46.5)

391 (31.4)

43 (55.1)

410 (51.8)

90 (4.4)

148 (3.5)

153 (15.5)

110 (8.8)

13 (16.7)

173 (21.8)

Cardiologist care Use of nonstatin lipid-lowering therapy Total number of medications taken* <5

1,107 (54.2)

2,628 (61.7)

164 (16.6)

308 (24.7)

19 (24.4)

115 (14.5)

5–9

557 (27.3)

1,022 (24.0)

342 (34.7)

491 (39.4)

23 (29.5)

310 (39.1)

$10

377 (18.5)

612 (14.4)

481 (48.7)

448 (35.9)

36 (46.2)

367 (46.3)

Data from the 30 days post-discharge Beta-blocker fill

1,827 (89.5)

3,913 (91.8)

817 (82.8)

1,123 (90.1)

69 (88.5)

662 (83.6)

Antiplatelet agent fill

1,505 (73.7)

3,622 (85.0)

700 (70.9)

1,052 (84.4)

51 (65.4)

602 (76.0)

Nonstatin lipid-lowering therapy fill Cardiac rehabilitation

59 (2.9)

97 (2.3)

87 (8.8)

34 (2.7)

2 (2.6)

80 (10.1)

323 (15.8)

880 (20.6)

142 (14.4)

254 (20.4)

9 (11.5)

107 (13.5)

Values are n (%). *In the 365 days prior to the index MI. †High-intensity statin is defined as atorvastatin 40 or 80 mg or rosuvastatin 20 or 40 mg. Simvastatin 80 mg is not included as a high-intensity statin. CHD ¼ coronary heart disease; CKD ¼ chronic kidney disease; MI ¼ myocardial infarction; PAD ¼ peripheral artery disease.

The percentage of beneficiaries >75 years of age

increased progressively between 2011 and 2014

whose first statin prescription fills following hospital

among U.S. adults. Among patients who filled statin

discharge for MI were for high-intensity doses

prescriptions, the percentage filling high-intensity

increased from 19.2% to 47.4% (Online Figure 3,

doses more than doubled over this time period, and

right). An increase in the percentage of beneficiaries

by the end of 2014, the majority of patients dis-

filling high-intensity statin prescriptions following

charged following hospitalization for MI filled high-

hospital discharge for MI occurred for beneficiaries

intensity doses. This trend was present among

initiating statin therapy (from 17.3% to 50.4%) and

younger adults with commercial health insurance and

those taking low- or moderate-intensity statins (from

older adults with Medicare government health in-

10.4% to 34.6%) and high-intensity statins (from

surance. The most commonly prescribed statin

82.7% to 87.8%) prior to their index MIs.

changed from simvastatin in the first quarter of 2011 to atorvastatin by 2014.

DISCUSSION

Several events occurred during the study period that may have contributed to the changes in high-

In our study, prescription fills for high-intensity

intensity statin use. In November 2011, the patent

statin therapy following hospital discharge for MI

for atorvastatin expired, making this agent available

2701

2702

Rosenson et al.

JACC VOL. 69, NO. 22, 2017 JUNE 6, 2017:2696–706

Trends in the Use of High-Intensity Statin Therapy

T A B L E 2 Characteristics of Medicare Beneficiaries, 66 to 75 Years of Age, Filling and Not Filling High-Intensity Statin Prescriptions

Following Hospital Discharge for Myocardial Infarction in 2014

Statin Initiators (n ¼ 8,975)

Prevalent Low- or ModerateIntensity Statin Users (n ¼ 8,105)

Prevalent High-Intensity Statin Users (n ¼ 2,569)

High-Intensity Statin† Fill

High-Intensity Statin Fill

High-Intensity Statin Fill

No (n ¼ 3,878)

Yes (n ¼ 5,097)

No (n ¼ 4,816)

Yes (n ¼ 3,289)

No (n ¼ 240)

Yes (n ¼ 2,329)

Age, yrs 66–70

1,974 (50.9)

2,716 (53.3)

2,341 (48.6)

1,698 (51.6)

106 (44.2)

1,229 (52.8)

71–75

1,904 (49.1)

2,381 (46.7)

2,475 (51.4)

1,591 (48.4)

134 (55.8)

1,100 (47.2)

2,127 (54.8)

3,047 (59.8)

2,664 (55.3)

2,008 (61.1)

122 (50.8)

1,412 (60.6)

Male Race/ethnicity White

3,397 (87.6)

4,433 (87.0)

4,001 (83.1)

2,765 (84.1)

196 (81.7)

1,964 (84.3)

African American

274 (7.1)

379 (7.4)

475 (9.9)

300 (9.1)

28 (11.7)

220 (9.4)

Asian American

40 (1.0)

64 (1.3)

91 (1.9)

66 (2.0)

4 (1.7)

30 (1.3)

Hispanic American

64 (1.7)

72 (1.4)

92 (1.9)

52 (1.6)

6 (2.5)

42 (1.8)

103 (2.7)

149 (2.9)

157 (3.3)

106 (3.2)

6 (2.5)

73 (3.1)

Low-income subsidy/dual eligible

Other

1,023 (26.4)

1,222 (24.0)

1,774 (36.8)

969 (29.5)

83 (34.6)

855 (36.7)

History of diabetes

1,069 (27.6)

1,198 (23.5)

2,607 (54.1)

1,559 (47.4)

148 (61.7)

1,328 (57.0)

History of CHD

648 (16.7)

567 (11.1)

2,349 (48.8)

1,146 (34.8)

123 (51.3)

1,509 (64.8)

History of stroke

100 (2.6)

87 (1.7)

331 (6.9)

153 (4.7)

18 (7.5)

172 (7.4)

History of heart failure

358 (9.2)

281 (5.5)

1,110 (23.0)

411 (12.5)

52 (21.7)

615 (26.4)

History of PAD

125 (3.2)

127 (2.5)

459 (9.5)

203 (6.2)

29 (12.1)

307 (13.2)

History of CKD

733 (18.9)

730 (14.3)

1,681 (34.9)

875 (26.6)

92 (38.3)

901 (38.7)

63 (1.6)

71 (1.4)

183 (3.8)

74 (2.2)

6 (2.5)

76 (3.3)

History of dementia Charlson comorbidity index 0

1,163 (30.0)

1,870 (36.7)

512 (10.6)

555 (16.9)

19 (7.9)

206 (8.8)

1–3

1,468 (37.9)

1,969 (38.6)

1,563 (32.5)

1,297 (39.4)

89 (37.1)

689 (29.6)

$4

1,434 (61.6)

1,247 (32.2)

1,258 (24.7)

2,741 (56.9)

1,437 (43.7)

132 (55.0)

Depression

467 (12.0)

475 (9.3)

886 (18.4)

451 (13.7)

43 (17.9)

457 (19.6)

Hospitalization in year prior to MI

662 (17.1)

671 (13.2)

1,635 (33.9)

730 (22.2)

74 (30.8)

838 (36.0)

Cardiologist care

871 (22.5)

827 (16.2)

2,439 (50.6)

1,279 (38.9)

135 (56.3)

1,456 (62.5)

Nonstatin lipid-lowering therapy*

322 (8.3)

286 (5.6)

707 (14.7)

377 (11.5)

48 (20.0)

505 (21.7)

<5

1,238 (31.9)

2,042 (40.1)

291 (6.0)

362 (11.0)

14 (5.8)

129 (5.5)

5–9

1,228 (31.7)

1,609 (31.6)

1,204 (25.0)

1,066 (32.4)

47 (19.6)

505 (21.7)

$10

1,412 (36.4)

1,446 (28.4)

3,321 (69.0)

1,861 (56.6)

179 (74.6)

1,695 (72.8)

Beta-blocker fill

3,246 (83.7)

4,455 (87.4)

3,690 (76.6)

2,715 (82.5)

183 (76.3)

1,761 (75.6)

Antiplatelet agent fill

2,371 (61.1)

3,432 (67.3)

2,744 (57.0)

2,195 (66.7)

131 (54.6)

1,389 (59.6)

Number of medications taken*

Data from the 30 days post-discharge

Nonstatin lipid-lowering therapy Cardiac rehabilitation

124 (3.2)

92 (1.8)

381 (7.9)

107 (3.3)

23 (9.6)

238 (10.2)

654 (16.9)

1,059 (20.8)

538 (11.2)

618 (18.8)

25 (10.4)

266 (11.4)

Values are n (%). *In the 365 days prior to the index myocardial infarction. †High-intensity statin is defined as atorvastatin 40 or 80 mg or rosuvastatin 20 or 40 mg. Simvastatin 80 mg is not included as a high-intensity statin. Abbreviations as in Table 1.

in a generic version to some pharmacy benefit plans.

in atorvastatin use began in 2011, which is consistent

By mid-2012, exclusive manufacturing rights expired,

with the timing of its generic availability.

resulting in more widespread and less costly generic

In June 2011, the U.S. Food and Drug Administra-

formulations of atorvastatin becoming available.

tion issued a black-box warning cautioning against

Although rosuvastatin 20 and 40 mg are considered

new prescriptions for simvastatin 80 mg because of

high-intensity statin therapy, fills for rosuvastatin

potential muscle toxicity, as reported in 2 clinical

declined following the generic availability of ator-

trials (15,16). The percentage of MarketScan and

vastatin. Previous studies have documented a tran-

Medicare beneficiaries who filled prescriptions for

sition from branded to generic medication after a

simvastatin 80 mg following hospital discharge for MI

drug comes off patent (14). In this study, the increase

declined more than 10-fold, and fewer than 1% of

Low/moderate intensity statin users Medicare

MarketScan

RR (95% CI) 0.97 (0.88, 1.07) 1.06 (1.00, 1.11) 1.04 (0.99, 1.08) 1 (reference)

RR (95% CI)

1 (reference) 0.98 (0.94, 1.01) 1.06 (1.02, 1.10)

1.09 (1.04, 1.13)

1.00

1 (reference) 0.97 (0.92, 1.02) 1.10 (1.04, 1.16) 1 (reference) 1.09 (0.99, 1.20) 1.10 (0.92, 1.31) 1.00 (0.80, 1.24) 0.98 (0.85, 1.14) 0.93 (0.87, 0.99) 1.00 (0.95, 1.07) 0.91 (0.85, 0.98) 0.91 (0.80, 1.04) 0.81 (0.74, 0.90) 1.01 (0.89, 1.15) 1.00 (0.93, 1.08) 0.89 (0.73, 1.08)

1.09 (0.99, 1.20) 0.85 (0.75, 0.96) 0.84 (0.61, 1.15) 1.18 (0.89, 1.58) 1.10 (0.82, 1.47) 0.78 (0.66, 0.93)

1 (reference) 0.98 (0.93, 1.03) 0.90 (0.79, 1.03) 1.04 (0.97, 1.12) 0.98 (0.90, 1.08) 0.97 (0.92, 1.02) 0.97 (0.87, 1.07)

1 (reference) 0.98 (0.93, 1.02) 0.94 (0.88, 1.00) 0.96 (0.89, 1.02) 1.01 (0.95, 1.08) 0.93 (0.88, 0.99) 0.91 (0.83, 1.00)

1 (reference) 0.95 (0.86, 1.04) 1.04 (0.86, 1.25) 0.96 (0.83, 1.10) 0.90 (0.77, 1.05) 0.88 (0.80, 0.96) 0.90 (0.77, 1.04)

1 (reference) 0.97 (0.90, 1.05) 0.93 (0.85, 1.02) 0.97 (0.89, 1.06) 0.90 (0.83, 0.97) 0.88 (0.83, 0.94) 1.13 (1.03, 1.23)

1 (reference) 0.94 (0.90, 0.98) 0.93 (0.87, 0.98)

1 (reference) 0.96 (0.92, 1.00) 0.92 (0.87, 0.98)

1 (reference) 0.96 (0.88, 1.05) 0.88 (0.80, 0.99)

1 (reference) 0.93 (0.86, 1.01) 0.87 (0.80, 0.95)

1.06 (0.99, 1.13) 1.28 (1.21, 1.35) 0.94 (0.83, 1.07) 1.09 (1.04, 1.13)

1.09 (1.03, 1.15) 1.11 (1.07, 1.15) 0.84 (0.71, 0.99) 1.06 (1.02, 1.11)

1.19 (1.04, 1.36) 1.40 (1.25, 1.57) 0.55 (0.41, 0.74) 1.09 (1.00, 1.18)

1.17 (1.09, 1.25) 1.23 (1.16, 1.30) 0.51 (0.42, 0.61) 1.21 (1.13, 1.28)

1.50

0.40 0.60

1.00

1.50

Relative Risk

0.40 0.60

1.00

1.50

Relative Risk

0.40 0.60

1.00

1.50

Relative Risk

Male sex, filling beta-blocker and antiplatelet agent prescriptions, and cardiac rehabilitation within 30 days following hospital discharge were associated with filling a high-intensity statin prescription in 2014 among MarketScan beneficiaries (left) and Medicare beneficiaries (right) initiating statin therapy and prevalent low- or moderate-intensity statin users prior to myocardial infarction (MI) hospitalization. CHD ¼ coronary heart

Rosenson et al.

Relative Risk

RR (95% CI)

1.12 (1.02, 1.23)

1 (reference) 1.08 (1.00, 1.15) 1.11 (0.94, 1.30) 0.96 (0.82, 1.13) 1.01 (0.91, 1.12) 0.99 (0.94, 1.03) 1.01 (0.96, 1.06) 0.92 (0.86, 0.99) 0.93 (0.80, 1.09) 0.90 (0.82, 1.00) 1.08 (0.94, 1.22) 0.97 (0.91, 1.03) 1.06 (0.90, 1.25)

1.03 (0.96, 1.09) 1.01 (0.94, 1.08) 1.02 (0.80, 1.31) 1.14 (0.93, 1.39) 1.14 (0.92, 1.41) 1.01 (0.91, 1.12)

0.40 0.60

Medicare RR (95% CI) 1.32 (1.06, 1.65) 1.05 (0.94, 1.18) 1.06 (0.98, 1.15) 1 (reference)

Trends in the Use of High-Intensity Statin Therapy

Age, years <40 40 – 49 50 – 59 60 – 64 66 – 70 71 – 75 Male Race/ethnicity White African American Asian American Hispanic American Other Low income subsidy/dual eligible History of diabetes History of CHD History of stroke History of heart failure History of PAD History of CKD History of dementia Charlson index 0 1-3 ≥4 Depression Hospitalization in year prior to MI Cardiologist care Use of non-statin lipid-lowering therapy Total number of medications taken <5 5–9 ≥10 Data from 30 days post-discharge Beta-blocker fill Antiplatelet agent fill Non-statin lipid-lowering therapy fill Cardiac rehabilitation

JACC VOL. 69, NO. 22, 2017

Statin initiators MarketScan

JUNE 6, 2017:2696–706

F I G U R E 2 Factors Associated With Filling High-Intensity Statins Following Hospitalization for Myocardial Infarction in 2014 Among MarketScan Beneficiaries <65 Years of Age and Medicare Beneficiaries

66 to 75 Years of Age, Initiating Statins and Who Were Taking Low- or Moderate-Intensity Statins Prior to Their Events

disease; CI ¼ confidence interval; CKD ¼ chronic kidney disease; PAD ¼ peripheral artery disease; RR ¼ relative risk. 2703

2704

Rosenson et al.

JACC VOL. 69, NO. 22, 2017 JUNE 6, 2017:2696–706

Trends in the Use of High-Intensity Statin Therapy

patients filled this medication in 2013 and 2014.

percentage of titration that occurs within 6 months of

Some, but not all, studies have reported black-box

hospital discharge, highlights the importance of

warnings to be associated with reductions in medi-

prescribing

cation use (16–18). Our study further highlights the

discharge following MI.

potential impact black-box warnings may have on drug prescriptions. In

November

high-intensity

statins

upon

hospital

An increase in high-intensity statin use was observed among MarketScan and Medicare benefi-

2011,

the

AHA/ACC

secondary

ciaries. However, high-intensity statin use following

prevention guidelines recommended that patients

MI

take the lowest statin dose needed to achieve a

compared with their counterparts in the MarketScan

low-density lipoprotein cholesterol level <70 mg/dl

database. This may be partially attributable to the

(6). Also, the ACC/AHA guideline on the treatment

age difference between these 2 cohorts and a greater

of blood cholesterol to reduce atherosclerotic CVD

burden of polypharmacy in older adults. High-

risk in adults was published in November 2013 (1).

intensity statin prescriptions were more likely to be

This latter guideline recommended high-intensity

filled among men and those who received beta-

statin use for patients #75 years of age with ACS

blockers and antiplatelet agents and attended car-

and moderate-intensity statins in those >75 years

diac rehabilitation following hospital discharge for

was

lower

among

Medicare

beneficiaries

of age with consideration of high-intensity statins

MI among both MarketScan and Medicare benefi-

on the basis of potential benefits for atheroscle-

ciaries. It is well documented that women are less

rotic CVD reduction versus the risk for adverse

likely than men to receive secondary prevention

reactions.

The

impact

of

the

2013

ACC/AHA

therapies following MI (23,24). Filling high-intensity

guideline on high-intensity statin use post-MI is

statin

not apparent from these results. Similar observa-

blockers and antiplatelet agents and attending car-

tions

evidence-based therapies across hospitals and providers or the use of intensive medical management

high-intensity statins in participants primarily with

for select high-risk patients (25,26). Patients filling

atherosclerotic CVD (19). It is possible that most

prescriptions for nonstatin lipid-lowering medica-

physicians were aware of the benefit of high-

tions in Medicare were less likely to fill prescriptions

intensity statins on the basis of the 2004 update

for high-intensity statins. One possible reason may

of the Adult Treatment Panel III guidelines (20)

be that these patients are intolerant to high-

and

(6).

intensity statins. Statin intolerance has been asso-

Also, we had only 1 year of follow-up after the

ciated with an increased risk for recurrent MI, sug-

publication of the 2013 ACC/AHA guideline, which

gesting the need for additional risk reduction

may not have provided sufficient time to observe

therapies (27).

impact

on

prevention

ACC

beta-

istry, which evaluated trends in moderate- and

full

the

with

diac rehabilitation may reflect variation in use of

secondary

from

conjunction

PINNACL

2011

reported

in

(Practice Innovation and Clinical Excellence) reg-

its

were

prescriptions

guidelines

high-intensity

statin

use

following hospital discharge for MI.

Strengths of the present study include the analysis of 2 large cohorts, 1 of younger patients with com-

Among patients filling low- or moderate-intensity

mercial health insurance and 1 of older patients with

statin prescriptions upon hospital discharge for MI,

government insurance. Using these 2 datasets, we

the percentage who were titrated to high-intensity

were able to investigate the use of high-intensity

statins within 6 months increased between 2011 and

statins across a broad age spectrum. Most U.S.

2014. However, 80% to 90% of patients in the

adults $65 years of age have health insurance

MarketScan and Medicare databases filling low- or

through Medicare, providing a high degree of gener-

moderate-intensity

alizability. The large sample size provided stable

statin

prescriptions

following

hospital discharge for MI in 2014 did not fill high-

estimates of high-intensity statin prescription fills.

intensity statin prescriptions within the next 6

STUDY LIMITATIONS. Despite these strengths, the

months. The reasons why few patients switch from

present study has known and potential limitations.

low- or moderate-intensity to high-intensity statins

We relied on pharmacy claims to identify statin use in

within 6 months of hospital discharge are unknown

the present analysis. However, substantial agreement

but may reflect previous intolerance, the presence of

between pharmacy claims and self-reported use and

comorbid conditions, or clinical inertia (21). Being

pill bottle review has been reported previously

hospitalized for MI represents a teachable moment,

(28,29). Patient behavioral and social support char-

and many patients may be amenable to treatment

acteristics and characteristics of the prescribing

changes

change

physician are not available in MarketScan and Medi-

following an MI, in conjunction with the low

care claims. Data were available only through 2014,

(22).

Patient

acceptability

to

Rosenson et al.

JACC VOL. 69, NO. 22, 2017 JUNE 6, 2017:2696–706

Trends in the Use of High-Intensity Statin Therapy

and there may have been an impact of the ACC/AHA

high-intensity statin use following hospital discharge

cholesterol guideline on high-intensity statin use in

for MI.

2015 and 2016. The present study relied on claims data, and we were unable to ascertain whether pre-

ADDRESS FOR CORRESPONDENCE: Dr. Robert S.

scription fills for low- or moderate-intensity statins

Rosenson, Mount Sinai Heart, Cardiometablomics

were appropriate on the basis of drug interactions,

Unit, Icahn School of Medicine at Mount Sinai, 1425

intolerance to high-intensity statins, or sufficient

Madison

control of low-density lipoprotein cholesterol. Data

York 10029. E-mail: [email protected].

Avenue,

MC1

Level,

New

York,

New

were available only for statin prescription fills and not prescriptions written. Race/ethnicity data are not

PERSPECTIVES

available in MarketScan. COMPETENCY IN PRACTICE-BASED LEARNING AND

CONCLUSIONS

IMPROVEMENT: High-intensity statin therapy prescriptions

The percentage of U.S. adults filling high-intensity statin prescriptions following hospital discharge for MI increased substantially between 2011 and 2014. This trend was present among younger commercially insured U.S. adults and older U.S. adults with government health insurance. Despite this favorable trend, a substantial percentage of patients filled lowor moderate-intensity statin prescriptions following hospital discharge for MI in 2014. The present study

after hospitalization for MI increased from 2011 to 2014, concurrent with the availability of generic atorvastatin. Prescription fills for high-intensity statins were associated with other good practices such as fills for antiplatelet and beta-blocker medications and attendance at cardiac rehabilitation. TRANSLATIONAL OUTLOOK: Further efforts are needed to increase high-intensity statin use following hospital discharge for MI.

highlights the need to continue efforts to increase

REFERENCES 1. Stone NJ, Robinson JG, Lichtenstein AH, et al. 2013 ACC/AHA guideline on the treatment of blood cholesterol to reduce atherosclerotic cardiovascular risk in adults: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol 2014;63:2889–934. 2. LaRosa JC, Grundy SM, Waters DD, et al. Intensive lipid lowering with atorvastatin in patients with stable coronary disease. N Engl J Med 2005;352:1425–35. 3. Cannon CP, Braunwald E, McCabe CH, et al. Intensive versus moderate lipid lowering with statins after acute coronary syndromes. N Engl J Med 2004;350:1495–504. 4. Rosenson RS, Kent ST, Brown TM, et al. Underutilization of high-intensity statin therapy after hospitalization for coronary heart disease. J Am Coll Cardiol 2015;65:270–7. 5. Hirsh BJ, Smilowitz NR, Rosenson RS, Fuster V, Sperling LS. Utilization of and adherence to guideline-recommended lipid-lowering therapy after acute coronary syndrome: opportunities for improvement. J Am Coll Cardiol 2015;66:184–92. 6. Smith SC Jr., Benjamin EJ, Bonow RO, et al. AHA/ACCF secondary prevention and risk reduction therapy for patients with coronary and other atherosclerotic vascular disease: 2011 update: a guideline from the American Heart Association and American College of Cardiology Foundation endorsed by the World Heart Federation and the

Preventive Cardiovascular Nurses Association. J Am Coll Cardiol 2011;58:2432–46. 7. Miller DR, Safford MM, Pogach LM. Who has diabetes? Best estimates of diabetes prevalence in the Department of Veterans Affairs based on computerized patient data. Diabetes Care 2004;27 Suppl 2:B10–21. 8. Rector TS, Wickstrom SL, Shah M, et al. Specificity and sensitivity of claims-based algorithms for identifying members of MedicareþChoice health plans that have chronic medical conditions. Health Serv Res 2004;39:1839–57. 9. Birman-Deych

E,

Waterman

AD,

Yan

Y,

Nilasena DS, Radford MJ, Gage BF. Accuracy of ICD-9-CM codes for identifying cardiovascular and stroke risk factors. Med Care 2005;43:480–5. 10. Tirschwell DL, Longstreth WT Jr. Validating administrative data in stroke research. Stroke 2002;33:2465–70. 11. Muntner P, Gutierrez OM, Zhao H, et al. Validation study of Medicare claims to identify older US adults with CKD using the Reasons for Geographic and Racial Differences in Stroke (REGARDS) Study. Am J Kidney Dis 2015;65:249–58. 12. Curtis LH, Whellan DJ, Hammill BG, et al. Incidence and prevalence of heart failure in elderly persons 1994–2003. Arch Intern Med 2008;168: 418–24. 13. Taylor DH Jr., Ostbye T, Langa KM, Weir D, Plassman BL. The accuracy of Medicare claims as an epidemiological tool: the case of dementia revisited. J Alzheimers Dis 2009;17:807–15.

14. Luo J, Seeger JD, Donneyong M, Gagne JJ, Avorn J, Kesselheim AS. Effect of generic competition on atorvastatin prescribing and patients’ out-of-pocket spending. JAMA Intern Med 2016;176:1317–23. 15. Armitage J, Bowman L, Wallendszus K, et al. Intensive lowering of LDL cholesterol with 80 mg versus 20 mg simvastatin daily in 12,064 survivors of myocardial infarction: a double-blind randomised trial. Lancet 2010;376:1658–69. 16. Forbes N, Cooray M, Al-Dabbagh R, et al. Domperidone prescribing practices exposed patients to cardiac risk despite a “black box” warning: a Canadian tertiary care center study. Can J Gastroenterol Hepatol 2016;2016:2937678. 17. Yu DT, Seger DL, Lasser KE, et al. Impact of implementing alerts about medication black-box warnings in electronic health records. Pharmacoepidemiol Drug Saf 2011;20:192–202. 18. Lasser KE, Seger DL, Yu DT, et al. Adherence to black box warnings for prescription medications in outpatients. Arch Intern Med 2006;166:338–44. 19. Pokharel Y, Tang F, Jones PG, et al. Adoption of the 2013 American College of Cardiology/ American Heart Association cholesterol management guideline in cardiology practices nationwide. JAMA Cardiol 2017;2:361–9. 20. Grundy SM, Cleeman JI, Merz CN, et al. Implications of recent clinical trials for the National Cholesterol Education Program Adult Treatment Panel III Guidelines. J Am Coll Cardiol 2004;44:720–32.

2705

2706

Rosenson et al.

JACC VOL. 69, NO. 22, 2017 JUNE 6, 2017:2696–706

Trends in the Use of High-Intensity Statin Therapy

21. Phillips LS, Branch WT, Cook CB, et al. Clinical inertia. Ann Intern Med 2001;135:825–34. 22. McBride CM, Emmons KM, Lipkus IM. Understanding the potential of teachable moments: the case of smoking cessation. Health Educ Res 2003; 18:156–70. 23. Cho L, Hoogwerf B, Huang J, Brennan DM, Hazen SL. Gender differences in utilization of effective cardiovascular secondary prevention: a Cleveland Clinic prevention database study. J Womens Health 2008;17:515–21. 24. Arnold SV, Spertus JA, Tang F, et al. Statin use in outpatients with obstructive coronary artery disease. Circulation 2011;124:2405–10. 25. Bradley EH, Herrin J, Mattera JA, et al. Quality improvement efforts and hospital

performance: rates of beta-blocker prescription after acute myocardial infarction. Med Care 2005;43:282–92. 26. Shafiq A, Valle J, Jang JS, et al. Variation in practice regarding pretreatment with dual antiplatelet therapy for patients with non-ST elevation myocardial infarction. J Am Heart Assoc 2016; 5:e003576.

assessing lipid-lowering medication use. Pharmacoepidemiol Drug Saf 2016;25:827–35. 29. Lauffenburger JC, Balasubramanian A, Farley JF, et al. Completeness of prescription information in US commercial claims databases. Pharmacoepidemiol Drug Saf 2013;22: 899–906.

27. Serban MC, Colantonio LD, Manthripragada A, et al. Statin intolerance and risk for recurrent myocardial infarction, coronary heart disease events and all-cause mortality following myocardial infarction. J Am Coll Cardiol 2017;69:1386–95.

KEY WORDS coronary artery disease, drug use, hydroxymethylglutaryl-CoA reductase

28. Colantonio LD, Kent ST, Kilgore ML, et al. Agreement between Medicare pharmacy claims, self-report, and medication inventory for

A PPE NDI X For supplemental tables and figures, please see the online version of this article.

inhibitors, secondary prevention