 |
|
Comparative Effectiveness of β-Blockers in Elderly Patients With Heart Failure
Judith M. Kramer, MD, MS;
Lesley H. Curtis, PhD;
Carla S. Dupree, MD, PhD;
David Pelter, BS;
Adrian Hernandez, MD;
Mark Massing, MD, PhD;
Kevin J. Anstrom, PhD
Arch Intern Med. 2008;168(22):2422-2428.
ABSTRACT
| |
Background Whether β-blockers (BBs) other than carvedilol, metoprolol succinate, and bisoprolol fumarate (evidence-based β-blockers [EBBBs]) improve survival in patients with heart failure (HF) is unknown. We compared the effectiveness of EBBBs vs non-EBBBs on survival.
Methods Our study population included North Carolina residents at least 65 years old who were eligible for Medicare and Medicaid with pharmacy benefits and had had at least 1 hospitalization for HF during the period 2001 through 2004. Primary outcome was survival from 30 days to 1 year. Secondary outcomes included number and days of rehospitalizations for HF and number of outpatient visits. Cohorts were defined by BB class (EBBBs, non-EBBBs, or no BBs) in first 30 days after discharge from index hospitalization for HF. Outcomes were analyzed using inverse probability–weighted (IPW) estimators with propensity score adjustment.
Results Of 11 959 patients, 40% were nonwhite, 79% were female, and 26% were at least 85 years old. Fifty-nine percent received no BB, 23% received EBBBs, and 18% received non-EBBBs. One-year adjusted mortality rates were 28.3% (no BBs), 22.8% (non-EBBBs), and 24.2% (EBBBs). The IPW-adjusted comparisons of 1-year mortality outcomes for either non-EBBBs or EBBBs compared with no BBs were statistically significant (P = .002 for both), but there was no statistical difference between the 2 BB groups (P = .43). The IPW-adjusted mean numbers of rehospitalizations for HF were 0.33 (no BBs), 0.29 (non-EBBBs), and 0.41 (EBBBs), with statistically more rehospitalizations in patients receiving EBBBs compared with no BBs (P = .002) and with non-EBBBs (P < .001).
Conclusion In this elderly population, the comparative effectiveness of EBBBs vs non-EBBBs was similar for 1-year survival, whereas the rehospitalization rate was higher for patients receiving EBBBs.
INTRODUCTION
In the past decade, randomized controlled trials have demonstrated that 3 β-blockers (BBs)—carvedilol (CoReg; GlaxoSmithKline, Philadelphia, Pennsylvania),1 metoprolol succinate (Toprol XL; Astra Zeneca, Wilmington, Delaware),2-3 and bisoprolol fumarate (Zebeta; Duramed Pharmaceuticals Inc, a subsidiary of Barr Laboratories Inc, Montvale, New Jersey)—and others4 improve survival in patients with heart failure (HF) characterized by left ventricular systolic dysfunction. These 3 drugs are considered evidence-based β-blockers (EBBBs) for treating HF with impaired systolic function.5 Several older, generic BBs (eg, atenolol, propranolol hydrochloride, and timolol maleate) have been shown in randomized trials to improve survival in patients who have experienced myocardial infarction (MI), but they have not been directly tested in patients with HF (and, thus, will be referred to henceforth as non–evidence-based β-blockers [non-EBBBs]).
Physicians are frequently confronted with a dilemma when patients being treated with a non-EBBB for hypertension or ischemic heart disease subsequently develop HF. The clinical dilemma is whether these patients need to be switched from the non-EBBB (which presumably is well tolerated but has an unknown effect on survival) to an EBBB (which would have an unknown adverse effect profile and likely a higher cost but which has a proven effect on survival). Observational data on patients with HF who, for various reasons, have continued treatment with non-EBBBs may provide useful information about the clinical outcomes associated with the use of these drugs. Using a large observational data set of a diverse population of patients with HF eligible for both Medicare and Medicaid (ie, dually eligible), we sought to determine the comparative effectiveness of EBBBs vs non-EBBBs on survival outcomes.
METHODS
STUDY POPULATION
North Carolina residents 65 years or older who were dually eligible for Medicare and Medicaid with drug benefits were considered for inclusion if they had had at least 1 hospitalization for HF during the period 2001 through 2004 and 6 months of claims data available prior to the index hospitalization for HF (Figure 1). Index hospitalization for HF was identified by the presence of a principal diagnosis of HF or by a principal cardiac diagnosis on a claim with a diagnosis of HF. Further exclusions are outlined inFigure 1.
|
|
|
|
Figure 1. Patient inclusions and exclusions. ESRD indicates end-stage renal disease; HF, heart failure; NC, North Carolina; SSN, Social Security number.
|
|
|
Treatment groups were determined by BB prescriptions filled in the first 30 days after discharge from the index hospitalization for HF. Patients with a pharmaceutical claim for carvedilol, metoprolol succinate or bisoprolol fumarate were classified as the EBBB group. Patients with a pharmaceutical claim for a BB other than carvedilol, metoprolol succinate, or bisoprolol fumarate were categorized as the non-EBBB group. Patients without any pharmaceutical claims for BBs were categorized as the no BB group. Because we reasoned that a patient receiving prescriptions for both an EBBB and a non-EBBB would derive benefit from the EBBB treatment, we classified such patients as receiving EBBBs for the purposes of the primary analysis. However, we also performed sensitivity analyses for all end points, excluding patients who received both EBBBs and non-EBBBs within the first 30 days.
The study population was identified as part of a Medicare quality improvement project conducted by the Carolinas Center for Medical Excellence, Inc (CCME) (Cary, North Carolina), the Medicare Quality Improvement Organization (QIO) for North Carolina and South Carolina. Data from Medicare were obtained by the CCME from the Centers for Medicare and Medicaid Services (CMS) under its QIO contract. North Carolina Medicaid data were provided to CCME from the Division of Medical Assistance (DMA) of the North Carolina Department of Health and Human Services. All data were housed and analyzed at CCME according to requirements specified in its business agreements with CMS and DMA.
OUTCOMES
The primary outcome measure was survival. Death dates were obtained from current Medicare beneficiary enrollment files with data current through December 31, 2004. Because treatment group assignment was determined by BB prescriptions filled in the first 30 days after discharge from index hospitalization for HF, we required a minimum of 30 days survival for inclusion in the analysis and analyzed only outcomes that occurred after the first 30 days. Secondary outcome measures included days survived, mean number of outpatient visits, mean number of rehospitalizations for HF, and mean number of days of rehospitalization for HF. For the secondary outcome measures involving rehospitalization for HF, we required a primary International Classification of Diseases, Ninth Revision (ICD-9) code of HF. As a sensitivity analysis, we compared outcome measures involving hospitalization for HF when the ICD-9 code was in any position.
STATISTICAL ANALYSIS
By treatment group, we summarized baseline characteristics as number (percentage) for categorical variables and as mean (SD) for continuous variables. We used the Pearson  2 test to assess differences in categorical variables and Wilcoxon rank sum tests for continuous variables. Baseline characteristics were adjusted using inverse probability–weighted (IPW) estimators corrected for the estimated propensity scores.6 Statistical significance was set at the P < .05 level.
To estimate the comparative effectiveness of EBBBs relative to no BB and non-EBBBs, we used IPW estimators7-9 with the data partitioned into monthly intervals. Events occurring within the first 30 days after discharge were not included in our analyses. Unadjusted estimates included weights based on Kaplan-Meier estimates of the treatment-specific censoring distributions. Adjusted estimates were inversely weighted by the product of estimated propensity scores and Cox proportional hazards regression model estimates of treatment-specific censoring distributions. Estimates of treatment effects, 95% confidence intervals (CIs), and P values were generated using SAS statistical software (version 8.2; SAS Inc, Cary, North Carolina) with robust standard errors.
We developed 3 propensity score models to adjust for treatment selection: (1) EBBBs vs all else, (2) non-EBBBs vs all else, and (3) no BB vs all else. Covariates in the treatment selection and Cox proportional hazards regression models included age (in 5-year increments), race, medications used within the 90 days prior to the index admission (non-EBBBs, EBBBs, aldosterone antagonists, angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, digitalis glycosides, inotropes, loop diuretics, nesiritide, thiazide diuretics, vasopressors), the Diagnostic Cost Group risk score,10-11 and individual comorbid conditions in the 6 months prior to the index hospitalization (history of MI, ischemic heart disease, hypertension, stroke, dementia, chronic obstructive pulmonary disease, diabetes mellitus, peripheral vascular disease, renal disease, and cancer).12-13 Year of admission was included in the treatment selection models only.
RESULTS
Table 1 compares the demographic characteristics of all North Carolina dually enrolled patients with drug coverage hospitalized with HF from 2001 through 2004 (N = 66 975) with the study population (N = 11 959). The study population was notably diverse: almost 40% were nonwhite, 79% were female, and 26% were 85 years or older. Compared with all dually enrolled patients from North Carolina with drug coverage hospitalized with HF during the time of the study, the study population included more women (79.3% vs 69.2%) and a higher percentage of patients older than 75 years. By study design, there were no patients younger than 65 years in the study population.
|
|
|
|
Table 1. Demographic Characteristics of All North Carolina Dually Enrolled Patients Hospitalized With HF From 2001 Through 2004 vs the Study Population
|
|
|
Baseline characteristics of the 3 treatment groups in the study population are shown in Table 2. Of the final analysis population, 7034 patients (58.8%) received no BBs, 2757 (23.1%) received an EBBB, and 2168 (18.1%) received a non-EBBB. The data in Table 2 reveal statistically significant imbalances in many baseline characteristics (eg, age, race, prior medications, underlying comorbidities, and year of admission) among the 3 treatment groups. Patients not prescribed BBs were older. A higher proportion of nonwhite patients (black and other) received EBBBs than did white patients. There was a trend to increasing use of EBBBs over time.
|
|
|
|
Table 2. Baseline Characteristics of Treatment Groupsa
|
|
|
After propensity score adjustment, baseline characteristics were well balanced between groups (Table 3). The only statistically significant difference in baseline characteristics after propensity score adjustment was prior treatment with loop diuretic. Use of a loop diuretic was not statistically different in the unadjusted analysis.
|
|
|
|
Table 3. Baseline Characteristics After Propensity Score Adjustmenta
|
|
|
MORTALITY
Outcomes at 1 year are displayed in Table 4. In both unadjusted and IPW-adjusted analyses, the death rate at 1 year was highest in the group receiving no BBs (28.3%, adjusted). In comparison with patients not prescribed BBs, patients receiving either a non-EBBB or an EBBB had a lower death rate (22.8% and 24.2%, adjusted, respectively). The differences in adjusted mortality between no BBs and non-EBBBs and between no BBs and EBBBs were 5.5 (95% CI, 2.6 to 8.4; P = .002) and 4.1 (95% CI, 1.7 to 6.6; P = .002), respectively. However, there was no statistically significant difference in death rate between patients taking non-EBBBs and EBBBs (difference: –1.4 [95% CI, –4.8 to 2.0]; P = .43). The number of days survived paralleled the death rates. Figure 2 shows IPW-adjusted cumulative mortality rates over the first year after discharge and demonstrates the higher mortality rate in patients receiving no BB compared with the 2 BB groups (non-EBBB group and EBBB group).
|
|
|
|
Table 4. Outcomes at 1 Year
|
|
|
|
|
|
|
Figure 2. Inverse probability–weighted (IPW), adjusted cumulative mortality rates over the first year after discharge. BB indicates β-blocker; EBBB, evidence-based β-blocker.
|
|
|
REHOSPITALIZATION
In both the unadjusted and IPW-adjusted analyses (Table 4), patients prescribed EBBBs had a statistically significant increase both in mean number of rehospitalizations for HF and in cumulative length of stay (mean days of rehospitalization for HF) compared both with patients receiving no BB and with patients prescribed non-EBBBs. In the adjusted analysis, compared with patients prescribed no BB, patients prescribed non-EBBBs had a statistically shorter length of stay and a trend toward fewer rehospitalizations for HF. The sensitivity analysis using a definition of hospitalization for HF with ICD-9 in any position was consistent with the analysis requiring a primary position for the ICD-9 code.
OUTPATIENT VISITS
Compared with patients prescribed no BB, those receiving either type of BB had a larger number of outpatient visits. In the IPW-adjusted analysis, there was no difference in the number of outpatient visits between patients receiving EBBBs and non-EBBBs. Sensitivity analyses excluding patients prescribed both EBBBs and non-EBBBs in the first 30 days showed qualitatively similar results for all end points.
COMMENT
In these diverse, elderly patients with HF, survival outcomes of patients prescribed non-EBBBs and EBBBs were similar. The plausibility of similar effects on survival associated with both non-EBBBs and EBBBs is enhanced by the findings of statistically significant (P = .002) improved survival for both categories of BBs compared with the group not treated with BBs. Our study also extends prior trial evidence by demonstrating these effects in an older, more diverse population than has been studied to date.
To our knowledge, the only randomized comparison of EBBBs vs non-EBBBs in patients with HF was the Carvedilol Or Metoprolol European Trial (COMET), which compared a maximum dose of carvedilol (EBBB) with a low dose of short-acting metoprolol (metoprolol tartrate, a non-EBBB).14 Although that study showed lower all-cause mortality for carvedilol, the fact that a low dose of metoprolol tartrate was tested has drawn criticism of the study.15
Observational data published from the Organized Program to Initiate Lifesaving Treatment in Hospitalized Patients with Heart Failure (OPTIMIZE-HF) registry16 described short-term mortality according to discharge use of BBs after a hospitalization for HF. (All patients had left ventricular systolic dysfunction.) Compared with patients who were not prescribed a BB, patients prescribed carvedilol at discharge had lower 60- to 90-day mortality rates. A secondary analysis produced similar results for metoprolol succinate and bisoprolol fumarate (both EBBBs). In an underpowered analysis (β = 0.3) of 386 patients prescribed non-EBBBs in the OPTIMIZE-HF registry, investigators found that the reduction in 60- to 90-day mortality by non-EBBBs compared with no BB (n = 382) was not statistically significant (HR 0.66; 95% CI, 0.37-1.17; P = .16).16 The direction and point estimate of this latter finding in the OPTIMIZE-HF registry is consistent with our findings, but, given the small sample size of patients prescribed non-EBBBs in the OPTIMIZE-HF registry and the short follow-up, the findings did not reach statistical significance. Of note, the OPTIMIZE study population was approximately 10 years younger than our population and predominantly male (63% vs 21% in our population).
The secondary end points in our study indicated that patients receiving EBBBs had statistically more rehospitalizations for HF and days of rehospitalization for HF than patients receiving non-EBBBs. The reasons for these findings are unclear. In an observational study such as ours, one always must be concerned about inadequate adjustment for confounders and unmeasured confounders that might explain such a finding. It is certainly possible that EBBBs were prescribed to patients who had increased risks for rehospitalization that we could not measure. It is also reasonable to consider that physicians who prescribe EBBBs might be more likely to provide better care and readmit for exacerbations. Finally, although there is no evidence to suggest it, there is the theoretical possibility that EBBBs could have adverse consequences that lead to repeated hospitalization but not death.
In contrast to our findings, in a recent study by Go et al,17 adjusted rates of rehospitalization for HF were not found to differ in over 7800 patients with HF receiving atenolol, short-acting metoprolol tartrate (both non-EBBBs), or carvedilol (EBBB). If a future study were to conduct a randomized survival comparison of EBBBs vs non-EBBBs, secondary outcomes including rehospitalization and outpatient visits should be tracked to quantify the overall resource impact of these alternative therapies. At a minimum, rehospitalization should be addressed in an observational data set that includes detailed clinical data.
Other findings were noteworthy. First, the high percentage of patients in our study prescribed no BBs (59%) is striking. Low prevalence of use of BBs in patients with HF has been described previously in the literature.18-19 Most paradoxical is research in Canada showing that patients with HF with impaired systolic function, who were at highest risk and most likely to benefit from BBs, were least likely to be prescribed BBs after a hospitalization for HF.20 Failure to prescribe BBs to patients with HF and failure of those patients to adhere to therapy mean that these patients forgo meaningful reductions in mortality and morbidity.21 In our particular study population, the low rate of use of BBs is difficult to interpret because the absence of information on left ventricular systolic function prevents us from estimating the proportion of the population with a guideline-based indication for BB therapy. However, the very low use of BBs in this vulnerable population of dually eligible patients would make this group an appropriate target for programs designed to implement such life-saving therapies.
Counterbalancing the reduction in mortality afforded by BBs in our study, we showed an increase in outpatient visits for either type of BB compared with no BB. The increase in clinic visits is consistent with the effect described in a single-center disease management program.22 However, the finding of an increased number of HF hospitalizations in our study contrasts with that single-center study of disease management and is inconsistent with the overall results of a meta-analysis of 19 studies of HF disease management by Whellan et al.23 In their meta-analysis,23 there was considerable heterogeneity of results across studies; none of the 4 studies using multiple centers demonstrated a reduction in hospitalizations. There was also limited information in that meta-analysis on the effect of the disease management programs on medication use.
Our study findings are consistent with a study of disease management across multiple centers (not included in the meta-analysis by Whellan et al23) in which a survival advantage was noted in patients with left ventricular systolic dysfunction, but no decrease in hospitalizations was found.24 Given that hospitalization accounts for a large part of the cost of care for HF, we were disappointed to find in our study that treatment with EBBBs is associated with an increase in the number of hospitalizations. However, if it were shown in randomized trials that non-EBBBs have survival effects similar to those of EBBBs, an economic benefit in reduced medication costs (owing to generic non-EBBBs) could be realized.
As an observational study, our results are subject to distortion by unmeasured confounders or residual confounding despite statistical adjustment. A rigorous test of the comparative effectiveness of EBBBs vs non-EBBBs would require a randomized survival study. We also had no information on left ventricular ejection fraction, an important prognostic factor and criterion in HF guidelines for treatment with BB.5 However, in this group of patients with HF with either preserved or reduced systolic function, the association of better survival with either type of BB (EBBBs or non-EBBBs) compared with no BB suggests the possibility that patients with preserved systolic function may also benefit from BB therapy.
We did not address the impact of nonadherence, switching to another BB after 30 days, or dosage. To the extent that some patients in the no-BB group actually received BBs as free samples or through a drug assistance program, the comparison of the no BB group with either the EBBB group or non-EBBB group may be biased toward the null.
CONCLUSIONS
In a population with HF that has not been studied to date—namely, elderly patients dually eligible for Medicare and Medicaid—we showed that treatment with either EBBBs or non-EBBBs compared with no BB is associated with a substantial survival benefit. Furthermore, we extended the evidence base by showing similar comparative effectiveness on 1-year survival between EBBBs vs non-EBBBs. Additional observational studies that compare the effectiveness of EBBBs with non-EBBBs in patients with HF with reduced left ventricular systolic function would be informative. If our results were to be replicated in such a population, the feasibility should be explored of conducting a randomized controlled trial in patients with HF comparing the effects of EBBBs and non-EBBBs on survival and rehospitalizations for HF.
AUTHOR INFORMATION
Correspondence: Judith M. Kramer, MD, MS, Duke Clinical Research Institute, PO Box 17969, Durham, NC 27715 (krame009{at}mc.duke.edu).
Accepted for Publication: May 13, 2008.
Author Contributions: Study concept and design: Kramer, Curtis, and Massing. Acquisition of data: Pelter and Massing. Analysis and interpretation of data: Kramer, Curtis, Dupree, Pelter, Hernandez, Massing, and Anstrom. Drafting of the manuscript: Kramer, Pelter, Massing, and Anstrom. Critical revision of the manuscript for important intellectual content: Kramer, Curtis, Dupree, Hernandez, and Massing. Statistical analysis: Pelter, Massing, and Anstrom. Obtained funding: Kramer. Study supervision: Kramer, Curtis, Dupree, and Massing.
Financial Disclosure: As executive director of the Clinical Trials Transformation Initiative (CTTI), a public private partnership that is one of the US Food and Drug Administration's Critical Path Programs, Dr Kramer received salary support in part from CTTI membership fees paid by the following organizations starting in the third quarter of calendar year 2008: Amgen Inc, Bayer, Biotronik, Bristol-Myers Squibb, C. R. Bard Inc, Eli Lilly and Company, Gemin X, Genentech, GlaxoSmithKline, Hoffman-La Roche Inc, Human Genome Sciences Inc, J&J Medical Devices & Diagnostics, J&J Pharmaceutical R&D, Pfizer, St Jude Medical, The Medicines Company, and Wright Medical. Dr Kramer received research funding from Pfizer and income for consulting from Icagen and Eli Lilly and Company. Dr Curtis received research funding from Allergan, GlaxoSmithKline, Eli Lilly and Company, Medtronic Inc, Novartis, Ortho Biotech, OSI Eyetech, and Sanofi-Aventis, and income for consulting from Allergan and Pfizer. Dr Hernandez received research funding from GlaxoSmithKline, Medtronic Inc, and Scios Inc, and income for consulting from Astra Zeneca, Novartis Pharmaceutical Co, and Thoratec Corp. Dr Anstrom received research funding from Alexion Inc, AstraZeneca, Bristol Myers Squibb, Eli Lilly and Company, Eyetech, Innocoll Pharmaceuticals, Medicure, Medtronic Inc, Medtronic Vascular, Novartis Pharmaceutical Company, Pfizer, and Procter & Gamble, and income for consulting from Pacific Pharmaceuticals.
Funding/Support: This work was supported in part by grant U18 HS 10548-07S1 from the Agency for Healthcare Research and Quality, US Department of Health and Human Services (DHHS), Rockville, Maryland. The analyses were performed at the CCME, the Medicare QIO for the state of North Carolina, under contract HHSM-500-2006-NC03C, sponsored by the CMS, DHHS.
Disclaimer: The content presented herein does not necessarily reflect the views or policies of the DHHS or the Centers for Medicare and Medicaid Services.
Additional Contributions: The North Carolina DHHS, Division of Medical Assistance, provided the pharmacy data used in these analyses. Amanda McMillan, MA, provided editorial assistance.
Author Affiliations: Department of Medicine, Duke Clinical Research Institute (Drs Kramer, Curtis, and Hernandez), and Department of Biostatistics and Bioinformatics (Dr Anstrom), Duke University Medical Center, and Duke Center for Education and Research on Therapeutics (Drs Kramer, Curtis, Anstrom, and Hernandez), Durham, North Carolina; Department of Medicine, University of North Carolina at Chapel Hill, Chapel Hill (Dr Dupree); and Carolinas Center for Medical Excellence, Cary, North Carolina (Drs Dupree and Massing and Mr Pelter).
REFERENCES
| |
1. Packer M, Bristow MR, Cohn JN; et al, U.S. Carvedilol Heart Failure Study Group. The effect of carvedilol on morbidity and mortality in patients with chronic heart failure N Engl J Med. 1996;334(21):1349-1355.
FREE FULL TEXT
2. Effect of metoprolol CR/XL in chronic heart failure: Metoprolol CR/XL Randomised Intervention Trial in Congestive Heart Failure (MERIT-HF). Lancet. 1999;353(9169):2001-2007.
FULL TEXT
|
WEB OF SCIENCE
| PUBMED
3. Hjalmarson A, Goldstein S, Fagerberg B; et al, MERIT-HF Study Group. Effects of controlled-release metoprolol on total mortality, hospitalizations, and well-being in patients with heart failure: the Metoprolol CR/XL Randomized Intervention Trial in congestive heart failure (MERIT-HF). JAMA. 2000;283(10):1295-1302.
FREE FULL TEXT
4. Cardiac Insufficiency Bisoprolol Study II (CIBIS-II). A randomised trial. Lancet. 1999;353(9146):9-13.
FULL TEXT
|
WEB OF SCIENCE
| PUBMED
5. 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 (Writing Committee to Update the 2001 Guidelines for the Evaluation and Management of Heart Failure). J Am Coll Cardiol. 2005;46(6):e1-e82.
FREE FULL TEXT
6. Lunceford JK, Davidian M. Stratification and weighting via the propensity score in estimation of causal treatment effects: a comparative study. Stat Med. 2004;23(19):2937-2960.
FULL TEXT
|
WEB OF SCIENCE
| PUBMED
7. Bang H, Tsiatis AA. Estimating medical costs with censored data. Biometrika. 2000;87:329-343.
FREE FULL TEXT
8. Anstrom KJ, Tsiatis AA. Utilizing propensity scores to estimate causal treatment effects with censored time-lagged data. Biometrics. 2001;57(4):1207-1218.
FULL TEXT
|
WEB OF SCIENCE
| PUBMED
9. Curtis LH, Hammill BG, Eisenstein EL, Kramer JM, Anstrom KJ. Using inverse probability-weighted estimators in comparative effectiveness analyses with observational databases. Med Care. 2007;45(10)(suppl 2):S103-S107.
PUBMED
10. Zhao Y, Ash AS, Ellis RP, Slaughter JP. Disease burden profiles: an emerging tool for managing managed care. Health Care Manag Sci. 2002;5(3):211-219.
FULL TEXT
| PUBMED
11. Ash AS, Posner MA, Speckman J, Franco S, Yacht AC, Bramwell L. Using claims data to examine mortality trends following hospitalization for heart attack in Medicare. Health Serv Res. 2003;38(5):1253-1262.
FULL TEXT
|
WEB OF SCIENCE
| PUBMED
12. Quan H, Sundararajan V, Halfon P; et al. Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data. Med Care. 2005;43(11):1130-1139.
FULL TEXT
|
WEB OF SCIENCE
| PUBMED
13. 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(5):480-485.
FULL TEXT
|
WEB OF SCIENCE
| PUBMED
14. Poole-Wilson PA, Swedberg K, Cleland JG; et al. Comparison of carvedilol and metoprolol on clinical outcomes in patients with chronic heart failure in the Carvedilol Or Metoprolol European Trial (COMET): randomised controlled trial. Lancet. 2003;362(9377):7-13.
FULL TEXT
|
WEB OF SCIENCE
| PUBMED
15. McBride BF, White CM. Critical differences among beta-adrenoreceptor antagonists in myocardial failure: debating the MERIT of COMET. J Clin Pharmacol. 2005;45(1):6-24.
PUBMED
16. Fonarow GC, Abraham WT, Albert NM; et al. Carvedilol use at discharge in patients hospitalized for heart failure is associated with improved survival: an analysis from Organized Program to Initiate Lifesaving Treatment in Hospitalized Patients with Heart Failure (OPTIMIZE-HF). Am Heart J. 2007;153(1):82.e1-e11.
PUBMED
17. Go AS, Yang J, Gurwitz JH, Hsu J, Lane K, Platt R. Comparative effectiveness of beta-adrenergic antagonists (atenolol, metoprolol tartrate, carvedilol) on the risk of rehospitalization in adults with heart failure. Am J Cardiol. 2007;100(4):690-696.
PUBMED
18. Smith NL, Chan JD, Rea TD; et al. Time trends in the use of beta-blockers and other pharmacotherapies in older adults with congestive heart failure. Am Heart J. 2004;148(4):710-717.
FULL TEXT
|
WEB OF SCIENCE
| PUBMED
19. Kramer JM, Chen AY, Hammill BG; et al. Time trends and factors associated with outpatient beta-blocker use in patients with heart failure or left ventricular dysfunction. Circulation. 2004;109:e231-e284.
FREE FULL TEXT
20. Lee DS, Tu JV, Juurlink DN; et al. Risk-treatment mismatch in the pharmacotherapy of heart failure. JAMA. 2005;294(10):1240-1247.
FREE FULL TEXT
21. Brophy JM, Joseph L, Rouleau JL. Beta-blockers in congestive heart failure: a Bayesian meta-analysis. Ann Intern Med. 2001;134(7):550-560.
FREE FULL TEXT
22. Whellan DJ, Gaulden L, Gattis WA; et al. The benefit of implementing a heart failure disease management program. Arch Intern Med. 2001;161(18):2223-2228.
FREE FULL TEXT
23. Whellan DJ, Hasselblad V, Peterson E, O'Connor CM, Schulman KA. Meta-analysis and review of heart failure disease management randomized controlled clinical trials. Am Heart J. 2005;149(4):722-729.
FULL TEXT
|
WEB OF SCIENCE
| PUBMED
24. Galbreath AD, Krasuski RA, Smith B; et al. Long-term healthcare and cost outcomes of disease management in a large, randomized, community-based population with heart failure. Circulation. 2004;110(23):3518-3526.
FREE FULL TEXT
 CiteULike  Connotea  Del.icio.us  Digg  Facebook  Reddit  Technorati  Twitter
What's this?
RELATED LETTERS
Eligibility Criteria for β-Blockade Might Have to Be Taken Into Account
Oscar M. P. Jolobe
Arch Intern Med. 2009;169(11):1077-1078.
EXTRACT
| FULL TEXT
Eligibility Criteria for β-Blockade Might Have to Be Taken Into Account—Reply
Judith M. Kramer, Lesley H. Curtis, Carla S. Dupree, Adrian F. Hernandez, Mark Massing, and Kevin J. Anstrom
Arch Intern Med. 2009;169(11):1078.
EXTRACT
| FULL TEXT
RELATED ARTICLES
Comparative Effectiveness of β-Blockers in Elderly Patients With Heart Failure—Invited Commentary
Brian L. Strom
Arch Intern Med. 2008;168(22):2428-2431.
EXTRACT
| FULL TEXT
Comparative Effectiveness of β-Blockers in Elderly Patients With Heart Failure—Invited Commentary
Bertram Pitt
Arch Intern Med. 2008;168(22):2431-2432.
EXTRACT
| FULL TEXT
THIS ARTICLE HAS BEEN CITED BY OTHER ARTICLES
|
Comparative effectiveness research: Relevance and applications to pharmacy
Schumock and Pickard
Am J Health Syst Pharm 2009;66:1278-1286.
ABSTRACT
| FULL TEXT
Eligibility Criteria for {beta}-Blockade Might Have to Be Taken Into Account--Reply
Kramer et al.
Arch Intern Med 2009;169:1078-1078.
FULL TEXT
Conflicts of Interest Need to Be Reported Reliably--Reply
Pitt
Arch Intern Med 2009;169:1079-1079.
FULL TEXT
Eligibility Criteria for {beta}-Blockade Might Have to Be Taken Into Account
Jolobe
Arch Intern Med 2009;169:1077-1078.
FULL TEXT
Conflicts of Interest Need to Be Reported Reliably
Whittle
Arch Intern Med 2009;169:1078-1079.
FULL TEXT
Correction: Which Beta-Blocker is Best for Patients with Heart Failure?
Journal Watch Cardiology 2009;2009:3-3.
FULL TEXT
Which Beta-Blocker is Best for Patients with Heart Failure?
Journal Watch Cardiology 2008;2008:2-2.
FULL TEXT
Comparative Effectiveness of {beta}-Blockers in Elderly Patients With Heart Failure--Invited Commentary
Strom
Arch Intern Med 2008;168:2428-2431.
FULL TEXT
Comparative Effectiveness of {beta}-Blockers in Elderly Patients With Heart Failure--Invited Commentary
Pitt
Arch Intern Med 2008;168:2431-2432.
FULL TEXT
|
