Influence of age on the management of heart failure: Findings from Get With the Guidelines–Heart Failure (GWTG-HF)

Congestive Heart Failure

Influence of age on the management of heart failure: Findings from Get With the Guidelines–Heart Failure (GWTG-HF) Daniel E. Forman, MD, a,b Christopher P. Cannon, MD, a Adrian F. Hernandez, MD, MHS, c Li Liang, PhD, d Clyde Yancy, MD, e and Gregg C. Fonarow, MD f for the Get With the Guidelines Steering Committee and Hospitals Boston, MA; Durham, NC; Dallas, TX; and Los Angeles, CA

Background Heart failure (HF) is common among elderly adults. Although multiple studies demonstrate age-related declines in the utilization of evidence-based therapies for coronary artery disease, there are few analyses of HF patients to distinguish possible age-related management differences. Methods

We analyzed 57,937 HF admissions from January 2005 through April 2007 in 257 hospitals participating in the American Heart Association's Get With The Guidelines–Heart Failure program. Patient characteristics and management were stratified by age groups ≤65, 66-75, 76-85, and >85 years. Multivariable regression analyses were used to assess the influence of age on use of therapies and inhospital mortality.

Results

The mean patient age was 73 ± 14 years; 18.7% were >85 years of age. Prescriptions of most HF therapies were relatively reduced with age but still remained high overall. Although 88.6% of patients ≤65 years of age with left ventricular systolic dysfunction were prescribed angiotensin-converting enzyme inhibitor or angiotensin receptor blocker and 90.9% were prescribed β-blockers, among those >85 years of age with left ventricular systolic dysfunction, 79% were prescribed angiotensin-converting enzyme inhibitor or angiotensin receptor blocker and 82.7% were prescribed β-blockers. Regression analysis that accounted for typical confounders demonstrated that older age was associated with diminished utilization of most evidence-based treatment measures as well as increased mortality.

Conclusions

Get With The Guidelines–Heart Failure data demonstrate that guidelines recommended therapies are frequently utilized for older patients with HF, including patients >85 years old. Nonetheless, age-related differences in therapy persist, suggesting that opportunities to improve care still remain. (Am Heart J 2009;157:1010-7.)

Optimal management of heart failure (HF) in older adults is a growing health care priority. Aging intrinsically predisposes to increased incidence of HF as well as to increased HF morbidity and mortality.1 Implications of these risks are compounded by today's burgeoning population of elderly.2 While HF is already the leading cause of hospitalization in persons ≥65 years old, cumulative morbidity, mortality, and associated costs will undoubtedly escalate as the elderly population increases.3,4

From the aCardiovascular Division, Brigham and Women's Hospital, Boston, MA, bVA Boston Healthcare System, Boston, MA, cCardiovascular Medicine, Duke University Medical Center, Durham, NC, dDuke Clinical Research Institute; Durham, NC, eBaylor Heart and Vascular Institute, Baylor University Medical Center, Dallas, TX, and fAhmansonUCLA Cardiomyopathy Center, University of California, Los Angeles, CA. Dr. Mihai Gheorghiade served as guest editor on this manuscipt. Submitted December 15, 2008; accepted March 4, 2009. Reprint requests: Daniel E. Forman, MD, Cardiovascular Division, Brigham and Women's Hospital, 75 Francis Street, Boston, MA 02115. E-mail: [email protected] 0002-8703/$ - see front matter © 2009, Mosby, Inc. All rights reserved. doi:10.1016/j.ahj.2009.03.010

Prior studies focusing on management of senior cardiovascular patients have highlighted the decreased use of evidence-based medicine in relation to advancing age, focusing predominantly on patients with coronary artery disease (CAD).5-7 However, it is often unclear whether these patterns stem from sound clinical judgment, such as concerns regarding frailty, or if they constitute inappropriate omissions.8 Confounding issues, including polypharmacy, iatrogenic side effects, comorbidities, and sex might also shape therapeutic choices. The American Heart Association's Get With The Guidelines-Heart Failure (GWTG-HF) program was initiated in January 2005 as part of a performance improvement initiative to enhance guideline adherence in patients hospitalized with HF.9-12 Institutional participation may help catalyze changes in therapeutic standards as clinicians become more cognizant of guideline-based recommendations.10,11 The prospectively collected GWTG-HF dataset also provides detailed clinical information on comorbidities and specific contraindications to therapy and therefore facilitates an important opportunity to assess practice patterns pertaining to younger versus older HF patients.

American Heart Journal Volume 157, Number 6

Methods Data collection Data were abstracted from the GWTG-HF from January 2005 through April 2007. This national, observational, prospective, and ongoing quality improvement initiative and data collection is supported in part by an unrestricted educational grant from GlaxoSmithKline. The GWTG-HF program and its component data elements have been described previously for the program's predecessor, OPTIMIZE-HF.12 The GWTG-HF participating hospitals include teaching and nonteaching, rural and urban, and large and small hospitals from all census regions of the United States. Hospitals participating in the registry submit clinical information online using an interactive case report form (Outcome Sciences, Inc, Cambridge, MA). Medical history, hospital care, and outcomes are assessed using consistent categories and standards. The GWTG-HF registry enrolls adults hospitalized with new or worsening HF as the primary reason for admission or with significant HF symptoms that developed during a hospitalization in which HF becomes the primary discharge diagnosis. Participating institutions submit consecutive eligible patients to the GWTG-HF database in compliance with Joint Commission and Centers for Medicare and Medicaid standards. All participants are required to comply with local regulatory and privacy guidelines and submit program protocols for review and approval by their institutional review board. Because data are used primarily for quality improvement, sites were granted a waver of informed consent under the common rule. The Duke Clinical Research Institute serves as the data analysis center, monitoring data for its completeness and accuracy, and analyzing aggregate deidentified data for research purposes.

Study population Sixty-one thousand three hundred four HF admissions from 257 hospitals participating in GWTG-HF were entered into its registry over the study period; 3,367 hospitalizations were excluded—1,428 because HF was not the primary reason for hospitalization and 1,939 because sex was not documented. Data were stratified into 3 age groups (≤65, 66-75, 76-85, and >85 years) based on Medicare eligibility and tiers of progressive age-related physiological vulnerability.13

Outcome measures The main outcome measures used were the guideline recommended therapies at discharge: • Complete discharge instructions • Evaluation of left ventricular ejection fraction (LVEF): before arrival, during hospitalization, or planned after discharge. • Angiotensin-converting enzyme inhibitor (ACEI) or angiotensin receptor blocker (ARB) for HF patients with left ventricular systolic dysfunction (LVSD), that is, LVEF b40%. • Adult smoking cessation advice/counseling for patients who smoked. • β-Blocker use at discharge for patients with LVSD without contraindications or intolerance.

Forman et al 1011

Additional quality metrics of interest were also assessed at discharge: • • • • •

Anticoagulant at discharge for patients with atrial fibrillation. Aldosterone antagonists for patients with LVSD. Hydralazine-nitrates in African-American patients with LVSD Blood pressure b140/90 mm Hg at discharge. Implantable cardioverter-defibrillator (ICD) placed or planned for patients with ejection fraction (EF) ≤30%

The first 4 performance measures comprise those advanced by the joint commission,14 used and publicly reported by federal agencies such as the Centers for Medicare and Medicaid Services.13 The second and third performance measures were parts of the Health Quality Alliance process measures.15 Although the fifth performance measure is not currently a national performance measure, recent evidence suggests it should be considered.11 Quality metrics were constructed using the numerator and denominator definitions defined by clinical performance measures for adults with HF16 and Joint Commission ORYX specifications14 assessing use rates among eligible patients without documented contraindications, intolerance, or other physician documentation. In these assessments, prescriptions of ACEI/ARBs, β-blockers, aldosterone blockers, and hydralazinenitrates were assessed only for those with LVEF b40%. Implantation of ICD was assessed only for those with EF ≤30%. The quality measures for thorough discharge instructions, smoking cessation counseling, blood pressure control, lipidlowering medications, and anticoagulation for atrial fibrillation were applied irrespective of LVEF. GWTG-HF also established two composite measures with which to assess treatment efficacy: • An opportunity quality of care index based on the number of therapeutic interventions in relation to cumulative circumstances when those interventions were indicated for the 5 GWTG-HF measures; a patient's composite adherence score was calculated as the sum of correct care provided divided by the total number of eligible opportunities (based on the 5 measures) for each patient. • An all or none measure for 100% compliance (%) on patients whether received 100% of evidence-based therapy for which they were eligible, up to a maximum of all 5 measures (all or none as an aggregate score), that is, this binary score excludes metrics from the denominator if patients had a contraindication or intolerance to a given therapy. Hospital lengths of stay and inhospital mortality were also assessed.

Statistical analysis Because age category is an ordinal variable, JonckheereTerpstra statistics were used to test the trend over age for continuous patient baseline characteristics variables and Cochran-Mantel-Haenszel row-mean scores statistics for categorical variables. Percentages were reported to describe distributions for categorical variables and medians with interquartile ranges for continuous variables. Multivariable logistic regression analyses, using the Generalized Estimating Equations method17 to adjust for clustering within hospitals, were performed to determine whether age independently

American Heart Journal June 2009

1012 Forman et al

Table I. Patient characteristics

Characteristic Female (%) Race (%) Caucasian African American Medicare payment (%) Ischemic HF (%) LVEF (%) Median (I-Q Range) LVEF b40% (% total population) PMHx (%) Cardiac HF CAD/IHD HTN Atrial fibrillation Hyperlipidemia Other CRI Anemia Pulmonary Dz DM( (noninsulin Rx) DM (insulin Rx) Alcohol abuse Tobacco Admission laboratory values (median, I-Q range) Cr (mg/dL) Hg (g/dL) Brain natriuretic peptide (pg/mL) Troponin (ng/mL)

Total Population

Age ≤65 y

Age 66-77 y

Age 76-85 y

Age>85 y

n = 57 937

n = 16 245 (28.0% total)

n = 12 488 (21.6% total)

n = 18 398 (31.8% total)

n = 10 806 (18.7% total)

P

50.6

40.1

47.1

53.3

65.6

b.0001

71.1 16.2 66.7 38.2 38 (25-55) 43.3

52.3 32.4 32.1 31.9 30 (20-50) 55.9

70.2 15.6 76.6 44.9 36 (25-55) 46.0

80.5 8.3 81.3 42.5 40 (28-55) 38.6

84.2 6.0 82.4 32.5 45 (30-60) 29.2

b.0001 b.0001 b.0001 b.0001 b.0001 b.0001

23.8 39.4 58.2 24.6 29.6

24.3 32.1 58.8 12.6 28.0

24.3 45.3 60.2 24.3 35.8

23.3 44.1 58.0 30.9 32.0

23.1 35.7 55.5 32.2 20.6

.07 b.0001 b.0001 b.0001 b.0001

15.7 13.2 23.0 14.5 19.6 1.1 15.4

14.3 9.9 22.7 19.3 19.4 2.6 33.4

17.6 12.9 27.2 19.0 24.1 0.9 16.0

16.8 14.7 23.7 12.4 20.4 0.4 6.7

13.9 16.0 17.5 5.9 13.1 0.1 2.4

b.0001 b.0001 b.0001 b.0001 b.0001 b.0001 b.0001

1.3 (1.0-1.9) 12.0 (10.6-13.5) 839 (427-1660)

1.3 (1.0-1.9) 12.5 (10.9-14.1) 814 (384-1690)

1.4 (1.0-2.0) 12.0 (10.6-13.5) 815 (398-1650)

1.4 (1.0-1.9) 11.9 (10.5-13.2) 849 (453-1646)

1.3 (1.0-1.8) 11.8 (10.5-13.1) 872 (468-1648)

.0992 b.0001 b.0001

0.05 (0.03-0.11)

0.05 (0.03-0.10)

0.05 (0.03-0.11)

0.05 (0.03-0.11)

0.06 (0.03-0.12)

b.0001

influenced each quality of care measure and outcome. In addition to age, the regression model adjusted for sex, race, body mass index at admission, systolic blood pressure at admission, insurance status (Medicare, Medicaid, others [Veteran's Administration and private], and no insurance), and medical history (anemia, chronic obstructive pulmonary disease, atrial fibrillation, peripheral arterial disease (PAD), hypertension, dyslipidemia, diabetes, renal insufficiency, prior myocardial infarction, smoker, prior heart failure, stroke, depression, and ischemic history/etiology). The analysis also adjusted for hospital characteristics (region, teaching status, number of beds, and surgical-percutaneous transluminal coronary angioplasty [PTCA]-transplant capabilities). A P value of b.05 was considered significant for each test. All analyses were performed using SAS software (version 8.2, SAS Institute, Cary, NC). The authors are solely responsible for the design and conduct of this study, including all analyses, drafting, and editing of this article and its contents.

Results The study population consisted of 57,937 heart failure patients, age range 18 to 107, median age 73 ± 14 years

from 257 hospitals. Table I shows that age distributions: 28% ≤65, 22% 66 to 75, 32% 76-85, and 19% >85 years. Although about 51% of the total population was female, proportions of women were higher in the older age strata (40% females ≤65 vs 66% females >85 years; P b .0001). Proportions of whites were also higher in the older age strata (52% ≤65 versus 84% >85, P b .0001). Proportions of patients receiving Medicare also increased relative to age (P b .0001); b1.5% of patients ≥66 years of age lacked health insurance. Table I also demonstrates that prior HF, both ischemic and nonischemic, was widely prevalent among the study population. While proportion of those with atrial fibrillation and anemia increased progressively as age increased, most other conditions (CAD, CVA, PAD, hyperlipidemia, hypertension, chronic renal insufficiency, pulmonary disease, and diabetes) increased through 66 to 75 years of age and declined thereafter. Tobacco and alcohol use also decreased relative to age, declines that are consistent with a likely survival effect in those free from the cardiac and noncardiac comorbid conditions. Table II and Figure 1 show the prespecified GWTG quality measures across age strata. Although prescriptions

American Heart Journal Volume 157, Number 6

Forman et al 1013

Table II. Quality of care composite measures % receiving Rx Age ≤ 65 y Age 66-77 y Age-76-85 y Characteristic GWTG quality indicators (%) HF patients discharged home with 6 instructions: activity level, diet, discharge medicines, follow-up appointment, weight monitoring, what to do if symptoms worsen HF patients with documentation of left ventricular function HF patients with LVSD discharged on ACEI/ARB HF patients with smoking history discharge with smoking cessation HF patients with LVSD discharged on β-blocker Other GWTG quality measures (%) HF patients with chronic or recurrent atrial fibrillation discharged on Warfarin HF patients with LVSD discharged with aldosterone antagonist African-American HF patients with LVSD discharged on hydralazine and isosorbide dinitrate combination HF patients with systolic BP b140 mm Hg and diastolic BP b90 mm Hg HF patients with EF b30% had ICD, had ICD placed or discharge with ICD Quality of care composite performance indices Opportunity Composite Measure Index (mean [SD]) All or none measure for 100% compliance (%)

Age>85 y

n = 57 937

n = 16 245

n = 12 488

n = 18 398

n = 10 806

P

77.1

77.5

77.8

76.6

75.9

.0063

92.6 84.7 87.1

93.8 88.6 88.7

93.1 84.3 86.3

92.8 81.8 84.0

89.4 79.0 74.4

b.0001 b.0001 b.0001

88.8

90.9

88.4

88.0

82.7

b.0001

64.7

71.0

68.3

64.6

53.9

b.0001

24.5

28.5

24.7

20.5

17.7

b.0001

5.4

5.7

4.8

5.7

3.1

.2921

74.5

71.7

74.8

76.3

77.0

b.0001

30.6

31.2

36.6

30.0

11.5

b.0001

84.7 (26.7) 69.1

85.7 (24.5) 67.2

85.0 (25.7) 68.1

84.8 (27.0) 70.3

82.7 (30.6) 71.0

.2201 b.0001

of many principal evidence-based therapies were relatively lower as age advanced, usage was still substantial among older HF patients. Seventy-nine percent of HF patients >85 years of age with LVSD were prescribed ACEI/ARB, and 82.7% were prescribed β-blockers despite declines compared to younger age strata. Similarly high use of evidence-based management pertained to discharge instructions and even smoking cessation, with only slight decrements in relation to age. In fact, successful blood pressure control was achieved to a greater extent in the older age strata (P b .0001). Multivariate regression analyses showed age-related performance decrements persisted even after accounting for common confounding variables for most measures. Table III reports ORs in relation to GWTG quality measures across older age strata used in the regression model. Three ORs are reported (66-75 vs ≤65, 76-85 vs ≤65, >85 vs ≤65 years old). Decrements in adherence relative to these age strata were detected for LVEF documentation, ACEI/ARB, smoking cessation, β-blockers, warfarin in atrial fibrillation, aldosterone antagonist, and ICD implantation. Discharge instructions and blood pressure (BP) control showed increased adherence among older age groups. The 2 GWTG quality of care composite indices are also included in Table II. The Opportunity Composite Measure showed no significant age-related declines (85% in those 66-75, and 82.7% in those >85 years of age; P = .22), indicating that among those who were determined

to be eligible for any of 5 specific therapies, high proportions received the treatments across younger or older HF patients. Furthermore, the GWTG All or None Measure for 100% Compliance index increased with age (from 67.2% in those ≤65 to 68.1% in those 66 to 75 years old, and to 71% in those >85 years old; P b .0001), suggesting that among those who were determined to be eligible for treatment, older HF patients were more likely to get 100% compliance for aggregate therapy. Less than 10% of patients ≥66 had procedures of any type in the course of the HF hospitalization, including b2% among patients >85 years old. Table IV lists the total implants of ICDs, biventricular pacemakers, right ventricular (RV) pacemakers, and cardioversion procedures during the incident HF hospitalization. ICD implants decreased from 5.2% in those 66 to 75 years old to only 0.4% in those >85 years old. Likewise, cardiac resynchronization therapy device placement fell from 2.6% in those 66 to 75, to 1.7% in those 76 to 85, and 0.6% in those >85 years old. Although permanent right ventricle pacemaker implants (single-chamber RV pacing and RV atrial ventricular pacing) increased slightly relative to age across the study population, total RV pacing implants were still distinctively low. Table V demonstrates older HF patients faced higher inhospital mortality. Mortality rates were 5.3% in patients >85 years old compared to 1.6% among patients ≤65 years old (P b .0001). Consistently, inhospital cardiovascular mortality increased with age (P = .025),

American Heart Journal June 2009

1014 Forman et al

Figure 1

Age-related differences in GWTG-HF quality indicators.

including increases in sudden cardiac death, acute coronary syndromes (ACS), and worsened HF. Multivariate analyses showed that age was an independent risk factor for mortality after accounting for the effects of confounders (66-75 vs ≤65 years: OR 1.68, 95% CI 1.39-2.03; 76-85 vs ≤65: OR 1.92, 95% CI 1.58-2.32; >85 vs ≤65 years: OR 2.62, 95% CI 2.13-3.22) with P b .0001 in each age-related analysis).

Discussion This GWTG-HF study demonstrates modest age-related declines in respect to some but not all evidence-based HF therapies but also demonstrates high overall use of evidence-based HF therapies even in the oldest HF patients. In particular, there was remarkably high use of ACEI/ARB and β-blockers in HF patients with LVSD despite advanced age. Furthermore, an age-stratified analysis of a prespecified GWTG All or None 100% Compliance index showed agerelated increases in guideline-based HF regimens. Our analysis indicates that not only are HF therapies being used in broad application for the older patient community, but that with stringent delineation of eligibility criteria, a subgroup of patients with indications for treatment were likely to receive the full complement of evidence-based care despite advanced age. GWTG-HF is one of several ongoing efforts to scrutinize caregiving for cardiovascular diseases as a means to better encourage use of evidence-based, guideline-recom-

mended care.10-12,18,19 Notably, American College of Cardiology/American Heart Association guidelines adhere to the premise that pathophysiological HF processes (and treatments) are equivalent for adults and do not vary recommendations based upon age.20 Nonetheless, among elderly, confounding issues are widespread, with agerelated physiological changes, antecedent cardiovascular disease, and comorbidites typically exacerbating instability and even constituting causal features underlying HF among older adults.1,21 Therefore, our data are especially important in demonstrating that many components of conventional HF care were provided at high rates to older patients in hospitals that participated in the GWTG-HF program and were presumably well tolerated during the incident hospitalization. Although the recently published TIME-CHF trial22 showed no overall benefit of brain natriuretic peptide– guided up-titration of evidence-based medications for a subgroup of HF patients ≥75 years old, this trial was primarily focused on dose up-titration and not the utility of the medications themselves. In comparison, GWTG-HF is primarily focused on the use of specific therapies in relation to incident HF hospitalizations. Among older CAD patients, poor adherence to evidence-based standards has been disconcerting. In 1997, Krumholz et al5 reported only 45% prescription of ACEIs at hospital discharge for eligible Medicare HF patients. Likewise, in the EFFECT study, ACEI/ARB and β-blocker treatment rates at time of hospital discharge

American Heart Journal Volume 157, Number 6

Forman et al 1015

Table III. Odds ratios for GWTG quality indices relative to age stratifications Outcome

Total n

Variable (y)

Adjusted OR

Lower 95% CI

Upper 95% CI

P

Discharge instructions

39 453

LVEF documentation

48 726

ACEI/ARB in LVSD

19 126

66-75 vs ≤65 76-85 vs ≤65 >85 vs ≤65 66-75 vs ≤65 76-85 vs ≤65 >85 vs ≤65 66-75 vs ≤65 76-85 vs ≤65 >85 vs ≤65 66-75 vs ≤65 76-85 vs ≤65 >85 vs ≤65 66-75 vs ≤65 76-85 vs ≤65 >85 vs ≤65 66-75 vs ≤65 76-85 vs ≤65 >85 vs ≤65 66-75 vs ≤65 76-85 vs ≤65 >85 vs ≤65 66-75 vs ≤65 76-85 vs ≤65 >85 vs ≤65 66-75 vs ≤65 76-85 vs ≤65 >85 vs ≤65 66-75 vs ≤65 76-85 vs ≤65 >85 vs ≤65

1.06 1.01 1.02 1.05 1.04 0.77 0.87 0.78 0.65 0.92 0.75 0.37 0.74 0.69 0.46 0.93 0.77 0.49 0.87 0.74 0.66 0.65 0.77 0.37 1.06 1.06 1.11 0.94 0.66 0.22

0.98 0.92 0.90 0.95 0.92 0.66 0.75 0.68 0.55 0.78 0.62 0.28 0.65 0.58 0.37 0.80 0.65 0.41 0.78 0.66 0.55 0.45 0.55 0.15 0.97 0.97 0.99 0.84 0.58 0.16

1.15 1.10 1.15 1.17 1.19 0.90 1.00 0.90 0.76 1.08 0.90 0.49 0.85 0.82 0.56 1.08 0.90 0.58 0.97 0.83 0.79 0.94 1.10 0.88 1.16 1.15 1.25 1.05 0.75 0.29

.16 .87 .80 .35 .51 .0009 .06 .0007 b.0001 .31 .002 b.0001 .00003 .00002 b.0001 .35 .001 b.0001 .010 b.0001 b.0001 .02 .15 .02 .20 .18 .07 .28 b.0001 b.0001

Smoking cessation counseling

7 766

β-Blocker in LVSD

18 733

Warfarin in atrial fibrillation

9 165

Aldosterone Antagonist in LVSD

18 904

Hydralazine/isosorbide dinitrate for African Americans

4 291

Systolic BP b140 mm Hg and diastolic BP b90 mm Hg at discharge

33 910

ICD in EF ≤30%

13 794

Table IV. Procedures during the incident HF hospitalization

Biventricular pacemaker RV pacemaker ICD Cardioversion

Age ≤65 y

Age 66-77 y

Age 76-85 y

Age >85 y

n = 12 782

n = 9867

n = 14 527

n = 8512

P

2.1 0.5 5.5 1

2.6 0.6 5.2 0.9

1.7 0.8 2.8 0.8

0.6 0.8 0.4 0.3

b.0001 b.0001 b.0001 b.0001

for eligible, older, higher-risk HF patients were only 64% and 32%, respectively.23 More recently, Alexander et al6 demonstrated frequent omission of many key therapeutic options for ACS in elderly in a community database, despite clinical trial data suggesting that many elderly would benefit disproportionately.24 Although some of the differences between these studies and this analysis might be attributed to the fact that caregivers participating in GWTG-HF were exceptionally motivated to achieve evidence-based treatment standards, GWTG-HF may have also provided a context that served to reinforce compliance of providers to treatment standards. Notorious underuse of guidelines-based HF therapy has also been described in long-term care facilities. Mann and

Evans25 reported only 41% and 38% prescription of ACEI and β-blockers respectively (only 16% were prescribed both) in eligible HF patients (mean age of 83 years old). Improved inhospital management of cardiovascular disease may also help improve subsequent management in long-term care facilities.7 It is also notable that although age is associated with increased mortality, inhospital death rates are still generally low in this population even among the oldest patients. Indeed, in contrast to hospitalizations for ACS where inhospital mortality for very old adults remains disproportionately high,26 the HF hospitalizations studied in HF-GWTG were less likely to culminate in death. Relatively higher utilization of evidence-based therapy is likely contributing to these differences.27

American Heart Journal June 2009

1016 Forman et al

Table V. Hospital death and LOS

Characteristic Patient died (%) Cause of death (for patients who died) CVD Non-CVD Not classified LOS (d) (median, [interquartile range])

Total population

Age ≤65 y

Age 66-77 y

Age-76-85 y

Age>85 y

n = 57 937

n = 16 245

n = 12 488

n = 18 398

n = 10 806

P

3.3

1.6

3.1

3.8

5.3

b.0001

46.7 14.7 38.6 5.0 (3.0-8.0)

40.6 19.9 39.5 5.0 (3.0-7.0)

46.5 13.9 39.6 5.0 (3.0-8.0)

47.5 14.4 38.1 5.0 (4.0-8.0)

48.7 13.2 38.1 5.0 (4.0-8.0)

.0250

Despite the high prescription of many key HF therapies for older patients, our data show low use of procedures in relation to age, including only rare use of ICDs for older patients with LVEF ≤30%. Although one might assume that additional utilization of ICDs and other procedures would increase mortality benefits to the elderly,28 management is not certain. In many cases, ICD placement and other procedures may have been performed during a subsequent hospitalization, allowing time for stabilization. Patients might also have been offered ICDs but declined the procedure. Further studies are necessary to delineate use and impact of procedures on outcomes for older HF patients.

Limitations The lack of follow-up after discharge remains a limitation of our analysis. Although it seems reasonable to assume that maximal therapy would translate into better long term outcomes, further study is needed, particularly because the comorbidities and frailties associated with aging may eventually lead to negative effects from therapies that were initially well tolerated. Another limitation is that our assumptions regarding shifting caregiving attitudes towards elderly HF patients are only inferred from treatment patterns. Our methods did not directly assess caregivers motives and logic, a perspective which would have complemented our assessments and conclusions. Furthermore, although GWTG-HF represents an opportunity to study HF patients in a real-world setting, there are several limitations to a registry-based study that call for careful interpretation of results. Data were collected by medical chart review and are dependent upon the accuracy and completeness of documentation and abstraction. Data collection is also dependent on voluntary participation and may have led to a tacit selection bias such that conclusions may not be generalizable. Likewise, findings may not apply to hospitals that differ in patient characteristics or care patterns from GWTG-HF hospitals. Furthermore, because of the large number of patients in the registry, some small differences that may be of little clinical relevance have P values indicating a high degree of statistical significance. Contraindications and intolerance were as docu-

b.0001

mented in the medical record, but a proportion of patients reported to be eligible for treatment but not treated may have had contraindications or intolerance that were present but not documented. In addition, there were no direct measures of socioeconomic status in GWTG-HF; thus, the contribution of certain elements affecting similar or disparate health care can only be inferred but not proven with these data.

Conclusions Using data collected as part of the GWTG-HF quality improvement program, we demonstrated that although age was associated with modest decrements in most guidelines-based HF treatments, use of major evidencebased treatments (ie, ACEI/ARB and β-blockers) still remained high among elderly HF patients, even adults >85 years old. These patterns differ from prior studies showing critical reductions in evidence-based therapy in relation to age among ACS and HF patients and thereby suggest that clinicians may have become relatively more compliant with guidelines-based therapeutic recommendations for their older patients, particularly in the framework of a guidelines-assessment program.

Disclosures Daniel Forman: honoraria, AstraZeneca (Wilmington, DE). Chris Cannon: research support, Accumetrics (San Diego, CA), AstraZeneca, Bristol-Myers Squibb (New York, NY)/Sanofi (Bridgewater, NJ) Partnership, GlaxoSmithKline (Philadelphia, PA), Merck (Whitehouse Station, NJ), Merck/ScheringPlough (Kenilworth, NJ). Partnership; clinical advisor, Automedics Medical Systems (San Diego, CA). Adrian Hernandez: research support, GlaxoSmithKline, Johnson & Johnson (Scios, Langhorne, PA), Medtronic (Minneapolis, MN), Novartis (East Hanover, NJ), Honoraria, AstraZeneca, Novartis. Li Liang, no conflicts. Clyde Yancy: research support, Medtronic, compensation declined; research consultant, theheart. org (Montreal, Quebec, Canada). Gregg Fonarow: research support, consultant, and honoraria, GlaxoSmithKline, Pfizer (New York, NY), Merck, Novartis, Medtronic, AstraZeneca, Scios.

American Heart Journal Volume 157, Number 6

Forman et al 1017

References 1. Thomas S, Rich MW. Epidemiology, pathophysiology, and prognosis of heart failure in the elderly. Clin Geriatr Med 2007;23:1-10. 2. DeFrances CJ, Hall MJ, Podgornik MN. 2003 National Hospital Discharge Survey. Advance data from vital and health statistics; no. 359. Hyattsville (MD): National Center for Health Statistics; 2005. 3. Redfield MM, Jacobsen SJ, Burnett JC, et al. Burden of systolic and diastolic ventricular dysfunction in the community. JAMA 2003;289: 194-202. 4. Roger VL, Weston SA, Redfield MM, et al. Trends in heart failure incidence and survival in a community-based population. JAMA 2004;292:344-50. 5. Krumholz HM, Vaccarino V, Ellerbeck EF, et al. Determinants of appropriate use of angiotensin-converting enzyme inhibitors after acute myocardial infarction in persons > or = 65 years of age. Am J Cardiol 1997;79:581-6. 6. Alexander KP, Peterson ED. Treatment of non–ST-elevation acute coronary syndrome in the elderly: current practice and future opportunities. Am J Geriatr Cardiol 2006;l15:42-9. 7. Krumholz HM, Radford MJ, Ellerbeck EF, et al. Aspirin for secondary prevention after acute myocardial infarction in the elderly: prescribed use and outcomes. Ann Intern Med 1996;124:292-8. 8. Wang TY, Robinson LA, Ou FS, et al. Discharge antithrombotic strategies among patients with acute coronary syndrome previously on warfarin anticoagulation: physician practice in the CRUSADE registry. Am Heart J 2008;155:361-8. 9. Hernandez AF, Fonarow GC, Liang L, et al. Sex and racial differences in the use of implantable cardioverter-defibrillators among patients hospitalized with heart failure. JAMA 2007;298:1525-32. 10. Fonarow GC, Abraham WT, Albert NM, et al. Influence of a performance-improvement initiative on quality of care for patients hospitalized with heart failure. Arch Intern Med 2007;167: 1493-502. 11. Fonarow GC, Abraham WT, Albert NM, et al. Association between performance measures and clinical outcomes for patients hospitalized with heart failure. JAMA 2007;297:61-70. 12. Fonarow GC, Abraham WT, Albert NM, et al. Organized program to initiate lifesaving treatment in hospitalized patients with heart failure (OPTIMIZE-HF): rationale and design. Am Heart J 2004; 148:43-51. 13. Centers for Medicare and Medicaid Services. Overview of specifications for hospital compare measures. Available at: http://www.cms.hhs.gov/HospitalQualitylnits/downloads/ HospitalOverviewOfSpecs200512.pdf. Last accessed April, 2009. 14. Joint Commission on Accreditation of Healthcare Organizations. Specifications manual for National Hospital Quality Measures. Available at: http://www.jointcommission.org/

15.

16.

17. 18.

19.

20.

21.

22. 23. 24.

25. 26.

27.

28.

PerformanceMeasurement/PerformanceMeasurement/Historical +NHQM+manuals.htm. Last accessed April, 2009. Centers for Medicare and Medicaid Services. The Hospital Quality Alliance (HQA) ten measure “Starter Set”. Available at: http://www. cms.hhs.gov/HospitalQualitylnits/downloads/ HospitalStarterSet200512.pdf. Last accessed April , 2009. Bonow RO, Bennett S, Casey Jr DE, et al. ACC/AHA clinical performance measures for adults with chronic heart failure: a report of the American College of Cardiology/American Heart Association Task Force on Performance Measures. J Am Coll Cardiol 2005;46: 1144-78. Zeger SL, Liang KY. Longitudinal data analysis for discrete and continuous outcomes. Biometrics 1986;42:121-30. Matthews JC, Johnson ML, Koelling TM. The impact of patient-specific quality-of-care report cards on guideline adherence in heart failure. Am Heart J 2007;154:1174-83. Bhatt DL, Roe MT, Peterson EDL, et al. Utilization of early invasive management strategies for high-risk patients with non–ST-segment elevation acute coronary syndromes: results from the CRUSADE Quality Improvement Initiative. JAMA 2004;292:2096-104. Hunt SA. ACC/AHA 2005 guideline update for the diagnosis and management of chronic heart failure in the adult: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol 2005;46:e1-e82. Lakatta EG, Levy D. Arterial and cardiac aging: major shareholders in cardiovascular disease enterprises: part II: the aging heart in health: links to heart disease. Circulation 2003;107:346-54. Pristerer M, Buser P, Ricki H, et al. BNP-guided vs symptom-guided heart failure therapy. JAMA 2009;301:383-92. Lee DS, Tu JV, Juurlink DN, et al. Risk-treatment mismatch in the pharmacotherapy of heart failure. JAMA 2005;294:1240-7. Bach RG, Cannon CP, Weintraub WS, et al. The effect of routine, early invasive management on outcome for elderly patients with non– ST-segment elevation acute coronary syndromes. Ann Intern Med 2004;141:186-95. Mann JL, Evans TS. A review of the management of heart failure in long-term care residents. Consult Pharm 2006;21:222-8. Alexander KP Newby LK, Armstrong PW, et al. Acute coronary care in the elderly, Part II. ST-segment-elevation myocardial infarction: a scientific statement for healthcare professionals from the American Heart Association Council on Clinical Cardiology. Circulation 2007; 115:2570-89. Wang R, Mouliswar M, Denman S, et al. Mortality of the institutionalized old-old hospitalized with congestive heart failure. Arch Intern Med 1998;158:2464-8. Ermis C, Zhu AX, Vanheel L, et al. Comparison of ventricular arrhythmia burden, therapeutic interventions, and survival, in patients b75 and patients ≥75 years of age treated with implantable cardioverter defibrillators. Europace 2007;9:270-4.