National Trends in Outcomes Among Elderly Patients with Heart Failure

The American Journal of Medicine (2006) 119, 616.e1-616.e7

CLINICAL RESEARCH STUDY

AJM Theme Issue: Cardiology

National Trends in Outcomes Among Elderly Patients with Heart Failure Mikhail Kosiborod, MD,a Judith H. Lichtman, PhD, MPH,b Paul A. Heidenreich, MD,c Sharon-Lise T. Normand, PhD,d Yun Wang, PhD,e Lawrence M. Brass, MD,b,f,g Harlan M. Krumholz, MD, SMe,h,i,j a

Mid America Heart Institute of Saint Luke’s Hospital and the University of Missouri, Kansas City, Mo; bSection of Chronic Disease Epidemiology, Department of Epidemiology and Public Health, Yale University School of Medicine, New Haven, Conn; cVA Palo Alto Health Care System, and the Department of Medicine, Stanford University, Stanford, Calif; dDepartment of Health Care Policy, Harvard Medical School and the Department of Biostatistics, Harvard School of Public Health, Boston, Mass; eCenter for Outcomes Research and Evaluation, Yale-New Haven Hospital, New Haven, Conn; fDepartment of Neurology, Yale University School of Medicine, New Haven, Conn; gNeurology Service of the VA Connecticut Healthcare System, West Haven, Conn; hSection of Health Policy and Administration, Department of Epidemiology and Public Health, Yale University School of Medicine, New Haven, Conn; iRobert Wood Johnson Clinical Scholars Program, Yale University School of Medicine, New Haven, Conn and jSection of Cardiovascular Medicine, Department of Medicine, Yale University School of Medicine, New Haven, Conn. ABSTRACT PURPOSE: Despite dramatic changes in heart failure management during the 1990s, little is known about the national heart failure mortality trends during this time period, particularly among the elderly. The purpose of this study was to determine temporal trends in outcomes of elderly patients with heart failure between 1992 and 1999. SUBJECTS AND METHODS: We analyzed a national sample of 3,957,520 Medicare beneficiaries aged 65 years or more who were hospitalized with heart failure between 1992 and 1999, assessing temporal trends in 30-day and 1-year all-causemortality and 30-day and 6-month all-cause hospital readmission. In risk-adjusted analyses, mortality and readmission for each year between 1994 and 1999 were compared with the referent year of 1993. RESULTS: Crude 30-day and 1-year mortality decreased slightly (range for 1992-1999: 11.0%-10.3% and 32.5%-31.7%, respectively), whereas 30-day and 6-month readmission increased (10.2%-13.8% and 35.4%-40.3%, respectively). After risk adjustment, there was no change in 30-day mortality between 1993 and 1999 (eg, for 1999 vs 1993, odds ratio [OR] 1.01, 95% confidence interval [CI], 1.00-1.02). One-year mortality was lower in 1994 compared with 1993 (OR 0.91, 95% CI, 0.90-0.92), but data from subsequent years suggested no continuous improvement after 1994 (1999 vs 1993: OR 0.93, 95% CI, 0.92-0.94). Thirty-day readmission increased (1999 vs 1993: OR 1.09, 95% CI, 1.07-1.10), but there was no change in 6-month readmission (1999 vs 1993: OR 1.00, 95% CI, 0.99-1.01). CONCLUSION: We found no substantial improvement in mortality and hospital readmission during the 1990s among elderly patients hospitalized with heart failure. These findings suggest that recent innovations in heart failure management have not yet translated into better outcomes in this population. © 2006 Elsevier Inc. All rights reserved. KEYWORDS: Heart failure; Health services research; Trends.

Dr. Kosiborod was a fellow in the Robert Wood Johnson Clinical Scholars Program at Yale University School of Medicine during the time this work was conducted. This publication was supported by grant number 1 K01 DP000085-01 from the Centers for Disease Control and Prevention (CDC). Its contents are solely the responsibility of the authors and do not necessarily represent the official views of CDC. The Centers for Medicare and Medicaid Services (CMS) reviewed and approved the use of its data for this work, and approved submission of the manuscript; this approval is based on data use only, and does not

0002-9343/$ -see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.amjmed.2005.11.019

represent a CMS endorsement of or comment on the manuscript content. Neither CDC nor CMS played a role in the design and conduct of the study, or in the analysis and interpretation of the data. All authors had full access to the data in the study and take responsibility for the integrity of the data and the accuracy of the analysis. Requests for reprints should be addressed to Harlan M. Krumholz, MD, SM, Yale University School of Medicine, 333 Cedar Street, PO Box 208088, New Haven, CT 06520.8088. E-mail address: [email protected].

616.e2

The American Journal of Medicine, Vol 119, No 7, July 2006

The decade of the 1990s brought dramatic changes to the linking records from the referring (transfer out) and acunderstanding and management of heart failure. Studies cepting (transfer in) hospitals. provided new insight into the underlying pathophysiology, We excluded the records of Medicare beneficiaries emphasizing the importance of maladaptive neurohormonal who had crossovers in coverage between fee-for-service mechanisms.1,2 New therapies such as angiotensin-convertand health maintenance organization status during each ing enzyme (ACE) inhibitors and calendar year, as well as those beta-blockers were introduced, with less than 12 months of conand their efficacy was demontinuous Medicare fee-for-service CLINICAL SIGNIFICANCE strated in clinical trials.3-16 Inteenrollment, because information gration of these therapies into about comorbidities and hospital ● Among patients hospitalized with heart clinical guidelines17 and their readmission for these beneficiafailure during the 1990s, there was no adoption in clinical practice were ries was not included in the daimprovement in short-term mortality. anticipated to produce substantial tabase. For patients with multi● There was no substantial improvement in improvements in heart failure ple admissions within each outcomes. calendar year, only the first adlong-term mortality, and no improveHowever, relatively little is mission was considered, thus exment in hospital readmission rates. known about how these changes cluding other admission records ● Recent advances in management have impacted the outcomes of patients during that year. In total, not yet resulted in better outcomes in with heart failure in the United 1,551,569 records met one or this patient group. States during that period. Almore of the exclusion criteria, though several cohort studies sugyielding a final sample of gested that long-term mortality 3,957,520 patients. has been improving among patients with heart failure during the past 30 to 50 years,18,19 Covariates they did not focus on trends during the 1990s. Results from Demographic variables including age, sex, and race were the few studies using administrative databases were conobtained from the Medicare Enrollment Database, which flicting, with some suggesting overall improvement in longuses Social Security Administration data.25 Prior cardiovasterm mortality between the early and late 1990s20-22 and cular diagnoses (myocardial infarction, coronary artery byothers finding no mortality improvement during the same pass surgery, percutaneous transluminal coronary angioperiod.23 Furthermore, these studies analyzed relatively plasty, prior heart failure, or prior admission for heart small patient samples,18-23 were geographically limited,18-23 failure) and comorbidities (diabetes, hypertension, demendid not focus on the elderly,19,20,22,23 or evaluated outcome tia, chronic obstructive pulmonary disease, and Deyo cotrends in health systems outside of the United States.20,22 To morbidity index26) were determined using ICD-9-CM diagdate, no study has provided a national perspective on recent nostic codes from both the index heart failure admission and outcome trends in patients with heart failure. To address this hospitalizations during the 12-month period before the inissue, we determined trends in short- and long-term outdex admission. Because our data collection began in 1992, comes in a national sample of Medicare beneficiaries hoscardiovascular diagnoses and comorbidities for 12 months pitalized with heart failure between 1992 and 1999. before that time were not included in the database. Thus, in

METHODS Study Cohort The study sample was derived from the Centers for Medicare and Medicaid Services’ administrative records and included all fee-for-service Medicare beneficiaries aged 65 years or more who were hospitalized with heart failure in acute care facilities between 1992 and 1999. Cohorts were identified for each calendar year from the Centers for Medicare and Medicaid Services’ Inpatient Standard Analytic Files using International Classification of Diseases, Ninth Revision, Clinical Modification [ICD-9CM] codes24 for a principal discharge diagnosis of heart failure (402.01, 402.11, 402.91, 404.01, 404.11, 404.91, 428.x). All records were cross-linked with the Medicare Enrollment Database (denominator file). For patients who were transferred between acute care facilities during hospitalization, a single episode of care was constructed by

the multivariable models we used the data for 1993 to 1999, with 1993 being the referent year.

Outcomes The outcomes were hospitalization rate, mean hospital length of stay, proportion of patients discharged to nursing facilities, all-cause mortality at 30 days and 1 year after admission and 1-year after discharge, and all-cause hospital readmission at 30 days and 6 months after discharge.

Statistical Analysis Patients were stratified by each calendar year from 1992 to 1999. We compared baseline demographic and clinical characteristics of patients between calendar years using Cuzick’s nonparametric test for trend27 for categoric variables, t test for continuous variables with normal (Gaussian) distribution, and Wilcoxon rank-sum test for continuous variables with skewed (non-Gaussian) distribution. We cal-

19.8 36.6 11.4 36.4 5.0 36.7 44.4 30.7 AMI ⫽ acute myocardial infarction; CABG ⫽ coronary artery bypass graft surgery; COPD ⫽ chronic obstructive pulmonary disease.

19.0 36.2 10.7 35.7 5.2 35.9 42.4 30.2 18.7 35.1 10.0 35.0 5.1 34.7 41.3 29.9 18.1 34.4 9.2 34.2 5.2 33.8 39.7 29.4 15.8 33.0 6.7 32.2 4.5 31.4 34.8 27.1 14.3 31.9 4.9 31.0 3.8 29.4 30.1 24.8

17.4 33.7 8.2 33.3 5.1 32.9 37.8 28.8

79.9 59.1 84.5 79.8 58.9 85.2 79.7 58.7 85.6 79.6 58.6 85.7 79.5 58.6 85.7 79.4 58.2 85.8 79.3 57.7 85.0

Demographics Mean age (y) 79.3 Female 57.3 White 85.9 Hospitalization/comorbidities AMI 9.7 Heart failure 15.5 CABG 3.8 COPD 26.7 Dementia 2.8 Diabetes 26.2 Hypertension 26.6 Deyo score ⱖ 3 14.2

Characteristic

The demographic characteristics of the patients did not change substantially during the study period (1992-1999 [age 79.3-79.9 years; female, 57.3%-59.1%; white, 85.9%84.5%]). We found a considerable increase, however, in the proportion of patients with comorbidities such as diabetes (26.2%-36.7%), hypertension (26.6%-44.4%), and dementia (2.8%-5.0%). The proportion of patients with a Deyo comorbidity score of 3 or more also increased from 14.2% to 30.7% (Table 1). Crude rates of hospitalization for heart failure among Medicare beneficiaries remained similar throughout the study period (1992-1999: 1.66%-1.82%, Figure 1). The mean length of stay decreased from 7.3 to 5.5 days, whereas the proportion of patients discharged to nursing facilities increased markedly (11.7%-16.7%, Figure 2). All-cause mortality changed little (30-day mortality: 11.0%-10.3%; 1-year mortality: 32.5%-31.7%; Table 2). All-cause hospital readmission rates steadily increased both at 30 days (10.2%13.8%, P ⬍ .001) and 6 months (35.4%-40.3%, P ⬍ .001; Table 3). Analyses adjusting for demographic variables and comorbidities showed no reduction in 30-day mortality from 1993 to 1999 (Table 2). Mortality at 1 year was lower in subsequent years compared with 1993 (1994 vs 1993: OR 0.91, 95% CI, 0.90-0.92). This was most likely the result of transiently higher mortality in 1993 observed in crude analysis, and the data did not suggest continuous improvement in 1-year mortality after 1994 (Table 2). The odds of readmission at 30 days increased for each subsequent year compared with 1993 (Table 3). Although the odds of readmission at 6 months decreased slightly in 1994 compared with 1993, and reached nadir in 1995, they subsequently returned to levels consistent with 1993 (1999 vs 1993: OR 1.00, 95% CI, 0.99-1.01, Table 3).

Table 1

RESULTS

Baseline Characteristics of Medicare Beneficiaries Hospitalized with Heart Failure between 1992 and 1999

culated crude heart failure hospitalization rates for each calendar year as a ratio of fee-for-service Medicare beneficiaries hospitalized with heart failure over the total number of eligible Medicare beneficiaries (obtained from the Medicare Enrollment Database). The mean length of hospital stay and the proportion of patients discharged to nursing facilities were calculated for each index year. We compared crude mortality rates at 30 days and 1 year between the calendar years of 1992 and 1999 using the Pearson chi-square test. By using multiple logistic regression models, we compared 30-day and 1-year mortality for each year between 1994 and 1999 with the referent year of 1993, after adjusting for demographic factors, medical history, and comorbidities. We used similar analyses to compare crude and adjusted hospital readmission rates at 30 days and 6 months from 1992 to 1999 and 1993 to 1999. All models also adjusted for clustering of patients by hospital using hospital-specific random effect. We conducted analyses with SAS 8.02 (SAS Institute Incorporated, Cary, NC).

16.6 32.1 8.1 33.0 4.6 32.6 37.1 26.9

616.e3

79.5 58.4 85.4

Heart Failure Trends 1992 1993 1994 1995 1996 1997 1998 1999 Total (N ⫽ 483,560) (N ⫽ 509,549) (N ⫽ 509,245) (N ⫽ 510,529) (N ⫽ 505,661) (N ⫽ 507,986) (N ⫽ 436,257) (N ⫽ 494,733) (N ⫽ 3,957,520) % % % % % % % % %

Kosiborod et al

616.e4

Figure 1 Temporal trends in crude heart failure hospitalization rates among Medicare beneficiaries from 1992 to 1999.

DISCUSSION Major Findings In this analysis of national administrative Medicare data, we demonstrate that overall, little progress was made in improving outcomes for elderly patients hospitalized with heart failure during the 1990s. Specifically, there has been no change in short-term mortality and only minor improvement in long-term mortality, whereas short-term hospital readmission rates have increased. Although these results contrast remarkably with the 20% to 30% relative reduction in mortality and the 30% to 40% reduction in hospital readmission that were seen in clinical trials of ACE inhibitors and beta-blockers,7-10,12 they are not entirely unexpected. There are several likely explanations for our findings. As demonstrated by recent studies, less than 20% of elderly patients with heart failure meet enrollment criteria for ACE inhibitor and beta-blocker clinical trials.28,29 Because of the high30 and increasing burden

The American Journal of Medicine, Vol 119, No 7, July 2006 of multiple comorbidities, as was observed in our study, many elderly patients with heart failure have contraindications to ACE inhibitors and/or beta-blockers.28,29 Furthermore, approximately half of elderly patients with heart failure have preserved left ventricular ejection fraction,31,32 a condition in which the efficacy of ACE inhibitors and beta-blockers has not been definitively demonstrated. Even among “ideal candidates,” less than 70% of elderly patients with heart failure are prescribed ACE inhibitors.29 Several studies have revealed that use rates of ACE inhibitors leveled off during the mid-1990s, and did not improve from 1995 to 2000.33,34 Overall, less than 20% of elderly patients with heart failure were receiving both a beta-blocker and an ACE inhibitor in the late 1990s.34 Compliance with therapy is also an issue, even among those for whom the medications are appropriately prescribed.35 All of these factors likely contributed to the remarkable “diffusion gradient” between the impact of therapeutic innovations on outcomes in clinical trials and relative lack of improvement in outcomes observed in our study.

PRIOR STUDIES OF TEMPORAL TRENDS IN HEART FAILURE SURVIVAL AND HOSPITAL READMISSION The results of several studies that examined recent trends in heart failure outcomes have been conflicting. Prospective cohort-based data from both the Framingham Heart Study and the Rochester Epidemiology Project have demonstrated improvements in long-term heart failure survival during the 1990s compared with the period from 1950 to 1969 and 1979 to 1984, respectively.18,19 However, those studies predominantly contrasted mortality during the 1990s with that of earlier decades and did not focus on mortality trends within the 1990s. Thus, these analyses could not detect the

Figure 2 Temporal trends in length of hospital stay and proportion of patients discharged to nursing facilities among Medicare beneficiaries hospitalized with heart failure between 1992 and 1999.

Kosiborod et al Table 2

Heart Failure Trends

616.e5

Trends in Crude and Adjusted Mortality Rates

Crude Mortality (%)

Adjusted Mortality (OR, 95% CI)

Year

30-day

1-year

30-day

1992 1993 1994 1995 1996 1997 1998 1999

11.0 10.9 10.6 10.5 10.3 10.2 10.2 10.3

32.5 33.9 31.7 31.5 31.4 31.7 31.8 31.7

1.00 0.99 1.00 0.99 0.98 0.99 1.01

NA (referent) (0.98-1.00) (0.98-1.01) (0.97-1.00) (0.97-0.99) (0.97-1.00) (1.00-1.02)

1-year 1.00 0.91 0.91 0.91 0.92 0.93 0.93

NA (referent) (0.90-0.92) (0.90-0.92) (0.90-0.92) (0.92-0.93) (0.92-0.93) (0.92-0.94)

CI ⫽ confidence interval; OR ⫽ odds ratio.

lack of improvement in heart failure outcomes during the middle and later parts of the decade. Similar to the current study, a Canadian administrative database study of hospitalized elderly patients from 1992 to 2000 found no change in 30-day mortality rates.22 Although there was a statistically significant trend toward lower 1-year mortality in that study, close analysis of the data suggests that most of the mortality change occurred between 1992 and 1993, and subsequent changes in crude mortality were minimal, with only 0.3% absolute reduction between 1993 and 1999. By using a similar administrative database of hospitalized patients in Scotland, MacIntyre et al.20 reported a 17% to 26% relative risk reduction in 30-day mortality, and a 15% to 18% relative risk reduction in 1-year mortality from the mid-1980s to the mid-1990s. Similarly, analysis of Medicare data from Northeast Ohio showed a 15% relative risk reduction in 30-day and 1-year mortality between the early and late 1990s.21 Although it is not clear why these studies found more significant mortality improvements than those found in the current study, some possible reasons are inclusion of younger patients with less comorbidity,20 different time frames of analysis,20,21 and limited patient samples and geographic areas of analysis.20,21 Finally, the Resource Utilization Among Congestive Heart Failure study showed no improvement in 1-year mortality between the early and late 1990s in a mixed sample of inpatients and outpatients with heart failure.23 An increase in 30-day readmission rates accompanied by a marked decrease in hospital length of stay as seen in our study are consistent with the findings of other investigators.36-39 These findings, combined with an increasing proportion of patients being discharged to nursing facilities, raise concern that patients with heart failure who are in relatively unstable condition are being prematurely discharged to less well-monitored settings, resulting in higher short-term readmission rates. The hypothesis that declining in-hospital length of stay may be affecting short-term outcomes is supported by a recent study by Baker et al.21 That study of 23,505 Medicare beneficiaries hospitalized with heart failure in Ohio revealed that mean length of stay

declined steeply from 9.2 days to 6.6 days between 1991 and 1997. Over the same period of time, in-hospital mortality decreased from just more than 6% to 4%. However, most of this decline was negated by a sharp increase in postdischarge mortality from 2% to approximately 5%.

TRANSLATING MEDICAL PROGRESS INTO PRACTICE The lack of substantial improvement in outcomes seen in our study underscores the complexity of applying medical progress to actual practice. Future efforts to improve outcomes in elderly patients with heart failure will likely need to be directed toward several key objectives: improving use of and adherence to lifesaving therapies in patients who are eligible to receive them, generating knowledge about appropriate therapies in patients with heart failure with preserved systolic function, and applying a multidisciplinary approach to the care of complex patients with heart failure who have multiple comorbidities. In consideration of the extremely high mortality in this patient group, increasing emphasis on health status outcomes, including functional status and health-related quality of life, and using interventions designed to prevent hospitalizations, such as disease management, may be warranted.

Study Limitations Our study provides an important national perspective on recent outcome trends in elderly hospitalized patients with heart failure, but several important issues should be considered in interpreting the results. It is possible that recent innovations in heart failure management have kept many “healthier” patients with heart failure out of the hospital. This may have resulted in “sicker” patients with multiple comorbidities being hospitalized during the late 1990s. In fact, our results suggest that the proportion of patients with multiple comorbidities increased considerably between the early and late 1990s, resulting in a “sicker” population of patients being hospitalized over time. Although our results

Table 3 Trends in Crude and Adjusted Hospital Readmission Rates Crude Hospital Readmission (%)

Adjusted Hospital Readmission (OR, 95% CI)

Year

30-day 6-month 30-day

6-month

1992 1993 1994 1995 1996 1997 1998 1999

10.2 11.6 12.2 12.6 12.8 13.3 13.6 13.8

1.00 0.98 0.96 0.96 0.97 1.00 1.00

35.4 36.9 37.6 38.2 38.7 39.3 40.1 40.3

1.00 1.02 1.02 1.03 1.06 1.08 1.09

NA (referent) (1.00-1.03) (1.01-1.03) (1.01-1.04) (1.04-1.07) (1.07-1.09) (1.07-1.10)

NA (referent) (0.97-0.99) (0.95-0.97) (0.95-0.97) (0.97-0.98) (0.99-1.01) (0.99-1.01)

CI ⫽ confidence interval; OR ⫽ odds ratio; NA ⫽ not available.

616.e6 were adjusted for this increase in the burden of comorbidities between the early and late 1990s, our data did not include information about left ventricular ejection fraction, cause of heart failure, degree of renal insufficiency, and medication use rates. Thus, a possibility of residual confounding by these factors cannot be excluded. We relied on administrative Medicare billing codes and did not independently validate the diagnosis of heart failure; however, previous studies have demonstrated that the specificity of these ICD-9-CM diagnostic codes for heart failure is approximately 93% compared with medical record review.40 We included only hospitalized patients in our analyses, and the results of this study may not apply to outpatients with heart failure. We focused on all-cause mortality and hospital readmission, and did not examine cause-specific outcomes. Finally, we only evaluated elderly patients—the age group within whom the overwhelming and increasing majority of patients with heart failure are found41—and our results may not be applicable to younger populations.

The American Journal of Medicine, Vol 119, No 7, July 2006

11.

12.

13.

14.

15.

16.

SUMMARY Despite significant advances in heart failure management, our results show a lack of substantial improvement in outcomes among elderly patients hospitalized with this condition during the 1990s. These data underscore the complexity of translating medical progress into clinical practice and emphasize the importance of coordinated efforts designed to improve outcomes in this patient group.

References 1. Francis GS, Goldsmith SR, Levine TB, et al. The neurohumoral axis in congestive heart failure. Ann Intern Med. 1984;101:370-377. 2. Packer M. The neurohormonal hypothesis: a theory to explain the mechanism of disease progression in heart failure. J Am Coll Cardiol. 1992;20:248-254. 3. The CONSENSUS Trial Study Group. Effects of enalapril on mortality in severe congestive heart failure. Results of the Cooperative North Scandinavian Enalapril Survival Study (CONSENSUS). N Engl J Med. 1987;316:1429-1435. 4. The SOLVD Investigators. Effect of enalapril on survival in patients with reduced left ventricular ejection fractions and congestive heart failure. N Engl J Med. 1991;325:293-302. 5. The SOLVD Investigators. Effect of enalapril on mortality and the development of heart failure in asymptomatic patients with reduced left ventricular ejection fractions. N Engl J Med. 1992;327:685-691. 6. Cohn JN, Johnson G, Ziesche S, et al. A comparison of enalapril with hydralazine-isosorbide dinitrate in the treatment of chronic congestive heart failure. N Engl J Med. 1991;325:303-310. 7. Flather MD, Yusuf S, Kober L, et al. Long-term ACE-inhibitor therapy in patients with heart failure or left-ventricular dysfunction: a systematic overview of data from individual patients. ACE-Inhibitor Myocardial Infarction Collaborative Group. Lancet. 2000;355:1575-1581. 8. Garg R, Yusuf S. Overview of randomized trials of angiotensinconverting enzyme inhibitors on mortality and morbidity in patients with heart failure. Collaborative Group on ACE Inhibitor Trials. JAMA. 1995;273:1450-1456. 9. Heidenreich PA, Lee TT, Massie BM. Effect of beta-blockade on mortality in patients with heart failure: a meta-analysis of randomized clinical trials. J Am Coll Cardiol. 1997;30:27-34. 10. Doughty RN, Rodgers A, Sharpe N, MacMahon S. Effects of betablocker therapy on mortality in patients with heart failure. A system-

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

atic overview of randomized controlled trials. Eur Heart J. 1997;18:560-565. Lechat P, Escolano S, Golmard JL, et al. Prognostic value of bisoprolol-induced hemodynamic effects in heart failure during the Cardiac Insufficiency BIsoprolol Study (CIBIS). Circulation. 1997;96:21972205. Lechat P, Packer M, Chalon S, et al. Clinical effects of beta-adrenergic blockade in chronic heart failure: a meta-analysis of double-blind, placebo-controlled, randomized trials. Circulation. 1998;98:11841191. Packer M, Bristow MR, Cohn JN, et al. The effect of carvedilol on morbidity and mortality in patients with chronic heart failure. U.S. Carvedilol Heart Failure Study Group. N Engl J Med. 1996;334:13491355. Pfeffer MA, Braunwald E, Moye LA, et al. Effect of captopril on mortality and morbidity in patients with left ventricular dysfunction after myocardial infarction. Results of the survival and ventricular enlargement trial. The SAVE Investigators. N Engl J Med. 1992;327: 669-677. Waagstein F, Bristow MR, Swedberg K, et al. Beneficial effects of metoprolol in idiopathic dilated cardiomyopathy. Metoprolol in Dilated Cardiomyopathy (MDC) Trial Study Group. Lancet. 1993;342: 1441-1446. Di Lenarda A, De Maria R, Gavazzi A, et al. Long-term survival effect of metoprolol in dilated cardiomyopathy. The SPIC (Italian Multicentre Cardiomyopathy Study) Group. Heart. 1998;79:337-344. Hunt SA, Baker DW, Chin MH, et al. ACC/AHA guidelines for the evaluation and management of chronic heart failure in the adult: executive summary: a report of the American College of Cardiology/ American Heart Association Task Force on Practice Guidelines (Committee to Revise the 1995 Guidelines for the Evaluation and Management of Heart Failure). Developed in collaboration with the International Society for Heart and Lung Transplantation; endorsed by the Heart Failure Society of America. Circulation. 2001;104:29963007. Levy D, Kenchaiah S, Larson MG, et al. Long-term trends in the incidence of and survival with heart failure. N Engl J Med. 2002;347: 1397-1402. Roger VL, Weston SA, Redfield MM, et al. Trends in heart failure incidence and survival in a community-based population. JAMA. 2004; 292:344-350. MacIntyre K, Capewell S, Stewart S, et al. Evidence of improving prognosis in heart failure: trends in case fatality in 66 547 patients hospitalized between 1986 and 1995. Circulation. 2000;102:11261131. Baker DW, Einstadter D, Thomas C, Cebul RD. Mortality trends for 23,505 Medicare patients hospitalized with heart failure in Northeast Ohio, 1991 to 1997. Am Heart J. 2003;146:258-264. Lee DS, Mamdani MM, Austin PC, et al. Trends in heart failure outcomes and pharmacotherapy: 1992 to 2000. Am J Med. 2004;116: 581-589. McCullough PA, Philbin EF, Spertus JA, et al. Confirmation of a heart failure epidemic: findings from the Resource Utilization Among Congestive Heart Failure (REACH) study. J Am Coll Cardiol. 2002;39: 60-69. The International Classification of Diseases, 9th rev., 3rd ed., Clinical Modification: ICD-9-CM. Washington, DC: Government Printing Office (DHHS publication no. (PHS) 89-1260.); 1989. Fleming C, Fisher ES, Chang CH, et al. Studying outcomes and hospital utilization in the elderly. The advantages of a merged data base for Medicare and Veterans Affairs hospitals. Med Care. 1992; 30:377-391. Deyo RA, Cherkin DC, Ciol MA. Adapting a clinical comorbidity index for use with ICD-9-CM administrative databases. J Clin Epidemiol. 1992;45:613-619. Cuzick J. A Wilcoxon-type test for trend. Stat Med. 1985;4:87-90.

Kosiborod et al

Heart Failure Trends

28. Masoudi FA, Havranek EP, Wolfe P, et al. Most hospitalized older persons do not meet the enrollment criteria for clinical trials in heart failure. Am Heart J. 2003;146:250-257. 29. Masoudi FA, Rathore SS, Wang Y, et al. National patterns of use and effectiveness of angiotensin-converting enzyme inhibitors in older patients with heart failure and left ventricular systolic dysfunction. Circulation. 2004;110:724-731. 30. Braunstein JB, Anderson GF, Gerstenblith G, et al. Noncardiac comorbidity increases preventable hospitalizations and mortality among Medicare beneficiaries with chronic heart failure. J Am Coll Cardiol. 2003;42:1226-1233. 31. Kitzman DW, Gardin JM, Gottdiener JS, et al. Importance of heart failure with preserved systolic function in patients ⬎ or ⫽ 65 years of age. CHS Research Group. Cardiovascular Health Study. Am J Cardiol. 2001;87:413-419. 32. Masoudi FA, Havranek EP, Smith G, et al. Gender, age, and heart failure with preserved left ventricular systolic function. J Am Coll Cardiol. 2003;41:217-223. 33. Stafford RS, Radley DC. The underutilization of cardiac medications of proven benefit, 1990 to 2002. J Am Coll Cardiol. 2003;41:56-61. 34. Smith NL, Chan JD, Rea TD, et al. Time trends in the use of betablockers and other pharmacotherapies in older adults with congestive heart failure. Am Heart J. 2004;148:710-717.

616.e7 35. Struthers AD, MacFadyen R, Fraser C, et al. Nonadherence with angiotensin-converting enzyme inhibitor therapy: a comparison of different ways of measuring it in patients with chronic heart failure. J Am Coll Cardiol. 1999;34:2072-2077. 36. Stewart S, MacIntyre K, MacLeod MM, et al. Trends in hospitalization for heart failure in Scotland, 1990-1996. An epidemic that has reached its peak? Eur Heart J. 2001;22:209-217. 37. Hall RE, Tu JV. Hospitalization rates and length of stay for cardiovascular conditions in Canada, 1994 to 1999. Can J Cardiol. 2003; 19:1123-1131. 38. Polanczyk CA, Rohde LE, Dec GW, DiSalvo T. Ten-year trends in hospital care for congestive heart failure: improved outcomes and increased use of resources. Arch Intern Med. 2000;160:325-332. 39. Baker DW, Einstadter D, Husak SS, Cebul RD. Trends in postdischarge mortality and readmissions: has length of stay declined too far? Arch Intern Med. 2004;164:538-544. 40. Goff DC Jr, Pandey DK, Chan FA, et al. Congestive heart failure in the United States: is there more than meets the I(CD Code)? The Corpus Christi Heart Project. Arch Intern Med. 2000;160:197-202. 41. Kozak LJ, Owings MF, Hall MJ. National Hospital Discharge Survey: 2001 annual summary with detailed diagnosis and procedure data. Vital Health Stat 13. 2004:1-198.