Cost-Effectiveness of Eplerenone Compared With Placebo in Patients With Myocardial Infarction Complicated by Left Ventricular Dysfunction an
http://www.100md.com
循环学杂志 2005年第3期
Emory University, Atlanta, Ga (W.S.W., Z.Z., P.K.)
New England Research Institutes, Watertown, Mass (E.M.M.)
Mid America Heart Institute, Kansas City, Mo (J.A.S.)
Caro Research, Inc., Boston, Mass (J.C., J.I.)
University of Massachusetts, Worcester (R.G.)
Prudential Equity Group, Deerfield, Ill (J.T.)
Pfizer, Inc, New York, NY (R.W.)
University of Michigan, Ann Arbor (B.P.).
Abstract
Background— In the Eplerenone Post-Acute Myocardial Infarction Heart Failure Efficacy and Survival Study (EPHESUS), aldosterone blockade with eplerenone decreased mortality in patients with left ventricular systolic dysfunction and heart failure after acute myocardial infarction. The present study was performed to evaluate the cost-effectiveness of eplerenone compared with placebo in these patients.
Methods and Results— A total of 6632 patients with left ventricular systolic dysfunction and heart failure after acute myocardial infarction were randomized to eplerenone or placebo and followed up for a mean of 16 months. The coprimary end points were all-cause mortality and the composite of cardiovascular mortality/cardiovascular hospitalization. The evaluation of resource use included hospitalizations, outpatient services, and medications. Eplerenone was priced at the average wholesale price, $3.60 per day. Survival beyond the trial period was estimated from data from the Framingham Heart Study, the Saskatchewan Health database, and the Worcester Heart Attack Registry. The incremental cost-effectiveness of eplerenone in cost per life-year and quality-adjusted life-year gained compared with placebo was estimated. The number of life-years gained with eplerenone was 0.1014 based on Framingham (95% CI, 0.0306 to 0.1740), 0.0636 with Saskatchewan (95% CI, 0.0229 to 0.1038), and 0.1337 with Worcester (95% CI, 0.0438 to 0.2252) data. Cost was $1391 higher over the trial period in the eplerenone arm (95% CI, 656 to 2165) because of drug cost. The incremental cost-effectiveness ratio was $13 718 per life-year gained with Framingham (96.7% under $50 000 per life-year gained), $21 876 with Saskatchewan, and $10 402 with Worcester.
Conclusions— Eplerenone compared with placebo in the treatment of heart failure after acute myocardial infarction is effective in reducing mortality and is cost-effective in increasing years of life by commonly used criteria.
Key Words: cost-benefit analysis ; heart failure ; myocardial infarction
Introduction
One of the most serious and frequent consequences of acute myocardial infarction (AMI) is heart failure, which develops in 22% of men and 46% of women after an MI.1 The presence of heart failure in patients with AMI is associated with a 55% greater risk of dying and 2.15-times-greater risk of death or recurrent AMI at 30 days.2 Patients with AMI who present to the hospital with heart failure have longer hospital stays, higher readmission rates, and higher mortality rates during hospitalization and 6 months after discharge than those without heart failure.3,4 With estimated direct and indirect health expenditures for heart failure nearing $26 billion annually,1 cost-effective treatment strategies for this disease are needed.
Multiple therapeutic strategies have been used to prolong life and to decrease hospitalizations for heart failure, including diuretics, ACE inhibitors or angiotensin receptor blockers (ARBs), ;-blockers, resynchronization therapy, implantable cardiac defibrillators, and heart transplantation. Additionally, nonselective aldosterone blockade has been shown to reduce mortality in patients with chronic, severe heart failure when used with ACE inhibitors, diuretics, and sometimes digoxin.5 Recently, the Eplerenone Post-Acute Myocardial Infarction Heart Failure Efficacy and Survival Study (EPHESUS)6 demonstrated that selective aldosterone blockade with eplerenone significantly reduced mortality and morbidity in patients with left ventricular systolic dysfunction (LVSD) and heart failure after AMI who were receiving optimal medical therapies. This trial was the first to demonstrate incremental benefit of a therapeutic agent in addition to standard therapy (including an ACE inhibitor or ARB and a ;-blocker) for improving outcomes in patients with heart failure after AMI.
Although its clinical efficacy is clear, the question remains as to whether the added benefit of eplerenone is worth the added cost. This study presents a cost-effectiveness analysis to define the value of the use of eplerenone compared with placebo in patients with heart failure after an AMI.
Methods
EPHESUS Trial Design
EPHESUS, a double-blind, multicenter, randomized, placebo-controlled trial, has previously been described in detail.6 Briefly, 6632 patients were recruited from December 1999 to December 2001 at 671 centers in 37 countries. Patients were randomized to eplerenone or placebo from 3 to 14 days after AMI. Inclusion criteria included LVSD (documented ejection fraction 40%) and heart failure documented by pulmonary rales, venous congestion on chest x-ray, or the presence of a third heart sound. Patients with diabetes were required to have LVSD but were not required to have documentation of heart failure. Patients were randomized to eplerenone (25 mg/d) or placebo for 4 weeks, after which the dosage of eplerenone was increased to a maximum of 50 mg/d. Patients received standard optimal medical therapy, which could include ACE inhibitors or ARBs, diuretics, ;-blockers, statin therapy, and coronary reperfusion. The 2 primary end points were time to death resulting from any cause and time to death caused by cardiovascular causes or first hospitalization for a cardiovascular event, including heart failure, recurrent AMI, stroke, or ventricular arrhythmia. The major secondary end points included death resulting from cardiovascular causes and death resulting from any cause or any hospitalization.
Economic Analysis and Costs
The economic analytic plan of EPHESUS was to compare the costs of the 2 treatment arms and, if the eplerenone arm was more costly and more effective, to perform an incremental cost-effectiveness analysis.7 Although not all sources of costs could be accounted for, the overall perspective was societal. Costs included in the analysis were direct medical care costs for hospitalizations, outpatient procedures, and drugs.8,9 No data were available from EPHESUS that could be used to calculate indirect costs resulting from lost productivity. All costs used 2001 as the base year, except the cost of eplerenone, which was not marketed until 2004. Costs and life expectancy differences were discounted 3% annually. The analysis used American unit costs but used resource use information and clinical outcomes for all 6632 patients. Cardiovascular healthcare resource use associated with the index and all follow-up hospitalizations, outpatient diagnostic tests and procedures, and medications were recorded prospectively.
Using a predefined algorithm designed by the investigators from case report form resource use, an investigator blinded to treatment group assigned the initial and subsequent hospitalizations for patients enrolled in EPHESUS to a diagnosis-related group (DRG) as used in the Medicare program in the United States. Costs for each DRG were estimated from average Medicare reimbursement rates obtained from the Medicare Part A data file,10 and professional costs were calculated by percent share by DRG according to the method of Mitchell et al.11 An investigator blinded to treatment group coded outpatient procedures by current procedural terminology and assigned a cost based on the Medicare fee schedule. All medications were assigned a cost based on Redbook average wholesale price (AWP).12 Specifically, the AWP for eplerenone was $3.60. All medications were assumed to continue for the duration of time that each patient was followed up.
Utility was measured with the EQ-5D13,14 in a subset from English-speaking countries of 1792 patients at baseline, 1530 patients at 6 months, and 1123 patients at 12 months. Quality-adjusted life-years (QALYs) were then calculated by multiplying survival by utility. For patients with a missing utility score, the average utility for all patients with available scores by treatment arm was used to estimate utility. Utility after 12 months was carried forward by use of the 12-month value. Because utility was measured in only a minority of patients, the primary analysis of cost-effectiveness remained cost per life-year gained, with cost per QALY gained as a sensitivity analysis.
Lifetime cost-effectiveness ratios in terms of cost per life-year gained and cost per QALY gained were predicted from in-trial estimates of incremental costs, event rates (death), and estimates of lost life expectancy associated with those in-trial deaths obtained from 3 sources: the Framingham Heart Study,12,15 the Saskatchewan Health database,16 and the Worcester Heart Attack Registry.17,18 These 3 sources were used to estimate survival because no single source perfectly met these criteria. For the Saskatchewan and Worcester databases, data on 2543 and 1094 patients, respectively, with heart failure after an AMI were analyzed with fractional polynomials and piecewise regression to obtain death hazard functions over time.19 These functions were adjusted according to patient characteristics through the use of separate Cox proportional-hazards models derived from the same data. For patients who died during the trial, life-years lost were obtained by subtracting the in-trial survival times from estimated age- and sex-specific life expectancy estimates.12 Patients were considered to have 0 life-years lost if they survived during the trial period. Average life-years lost for each treatment group were calculated across all patients who died and survived in each arm of the trial. The difference in average life-years lost because of deaths (placebo minus eplerenone) yields an estimate of the life-years gained with eplerenone. Life-years and QALYs were discounted at 3% annually.12 The additional healthcare costs attributed to life-years gained by treatment were estimated as a sensitivity analysis.20 Costs beyond the trial period were estimated by calculating the costs during the trial by year, carrying forward the average cost in years 2 and 3 of the trial, and discounting by 3% annually. Bootstrap methods were used to estimate the fraction of the joint distribution of the cost and effectiveness differences lying in different regions of the cost-effectiveness plane.21,22
In addition to applying these methods to the overall population, we also performed these cost-effectiveness analyses for certain demographic subgroups defined by age, sex, diabetes, and prior AMI.
Results
There were no differences in the baseline characteristics of age, gender, prior AMI, diabetes, hypertension, prior history of heart failure, and ejection fraction (Table 1). There was a 15% relative decrease in death from any cause with eplerenone compared with placebo. There was a 13% relative decrease in the other primary end point of death or hospitalization for cardiovascular events and significant decreases in the secondary end points of death from any cause or any hospitalization, sudden death from cardiac causes, and the number of episodes of hospitalization for heart failure. The average follow-up was 16 months.
The incremental cost-effectiveness ratio (ICER) of eplerenone compared with placebo in the analysis with Framingham was $13 718 per life-year gained, with 96.7% of estimates falling below the threshold of $50 000 per life-year gained (Table 6). When the Saskatchewan estimates of life expectancy were used, the ICER was $21 876, with 93.8% of estimates under $50 000 per life-year gained. Based on Worcester estimates of life expectancy, the ICER was $10 402 per life-year gained, with 98.8% of observations falling below the threshold of $50 000 per life-year gained. The ICERs were systematically higher when calculated in cost per QALY gained. The ICERs were also systematically higher if the costs after the trial period were included. The joint bootstrap distribution of the difference in efficacy in life-years and cost is displayed in Figure 1, with the Framingham estimates for lost life expectancy used. Almost all estimates are in quadrant 1 of the cost-effectiveness plane, meaning that there was greater efficacy at increased cost with eplerenone. The diagonal line from the origin represents $50 000 per life-year gained. Estimates below this line in quadrant 1 or 2 would be cost-effective, if $50 000 represents society’s willingness-to-pay threshold.
Discussion
This analysis revealed that eplerenone in the setting of heart failure after an acute MI is a life-saving medication that is cost-effective compared with placebo by the common benchmark ceiling ratio of $50 000 per life-year gained. This conclusion was robust throughout a range of projections using 3 different sources for estimates of lost life expectancy resulting from in-trial deaths in a patient population in which the majority received both ;-blockers (75%) and ACE inhibitors or ARBs (87%).6 Furthermore, the bootstrap analysis revealed that with each method of costing, >90% of simulations were below the $50 000 benchmark.
Previous reports from EPHESUS documented improved survival and fewer cardiovascular events in patients experiencing heart failure after an AMI who are treated with eplerenone compared with placebo.6 The results from EPHESUS complement previous results from the Randomized ALdactone (spironolactone) Evaluation Study for congestive heart failure (RALES), which showed improved survival in patients with severe chronic heart failure who were treated with spironolactone and in whom recent MI was excluded.5 EPHESUS is the first and only study to demonstrate the efficacy of aldosterone blockade for reducing mortality and morbidity in post-AMI patients with heart failure. It is important to note that eplerenone was effective in preventing events in these patients who were already treated optimally with ;-blockers and either ACE inhibitors or ARBs and that this is the only agent proven to add incremental benefit on mortality and morbidity above and beyond standard therapy in these patients;6 ;-blockers and ACE inhibitors have been shown to be clinically effective and cost-effective in the treatment of heart failure.23 The present study reveals that compared with placebo, eplerenone is cost-effective in the treatment of post-MI heart failure in a population of patients already receiving standard therapy.
Study Limitations
The follow-up period in EPHESUS was of variable length (range, 0 to 33 months). Thus, cost-effectiveness was calculated for the average follow-up period of 16 months, with therapy with eplerenone for that period of time but not thereafter. This study does not and realistically cannot address the issue of how long eplerenone should be taken. The estimation of life expectancy assumes that the survival curves remain parallel after the trial period.
The in-trial results cannot give a full picture of the survival advantage of eplerenone. Thus, the estimation of survival was extended beyond the study period by using data from 3 separate sources. However, the degree to which the survival experience of patients from these observational studies yields accurate estimates of life expectancy for the EPHESUS population is uncertain. In both the Saskatchewan and Worcester databases, mortality within 1 year of MI was higher than in EPHESUS. This would mean a shorter projected life expectancy, rendering our results conservative. In addition, there was considerable variation between the estimates from these 3 sources, given variations in the populations. The Framingham Heart Study was used because it is a large, well-known, epidemiological database.15 The Worcester and Saskatchewan databases also were chosen because they included patients similar to those in EPHESUS (ie, post-AMI heart failure patients), and both databases included long-term data.16–18 Although the Worcester database had the greatest similarity to EPHESUS and the closest estimate of mortality hazard, the ICERs from these 3 databases were all in an acceptable range and provide a sensitivity analysis supporting the cost-effectiveness of eplerenone.
The costing methodology was based on Medicare payments for hospitalization, the Medicare fee schedule for procedures, and AWPs for medications. Eplerenone was priced at the AWP of $3.60 a day. The extent to which these costs reflect resource use from a societal point of view is somewhat uncertain because there is no single source for costs that unequivocally represents societal costs. Thus, Medicare payments may appropriately reflect costs for the Centers for Medicare and Medicaid Services but may not adequately represent resource consumption by hospitals and physicians because Medicare costs tend to be lower than managed care costs. Therefore, the present analysis may provide a conservative representation of the cost-effectiveness of eplerenone.
Costing beyond the trial period was based on projections of costs within the trial period. This approach, while reasonable, is not easily subject to empirical assessment.
Indirect costs were not included, especially those reflecting return to work. To the extent that eplerenone permits return to work by preventing hospitalizations or deaths, this would lower costs in the eplerenone arm, again rendering the results conservative.
Resource use and clinical outcomes were considered trial wide. Use of Medicare costing does not account for possible within-DRG differences in treatment practices and resource use across countries. With so many countries, it is not possible to adequately account for variation in DRGs across countries. Thus, the use of trial-wide data for costing in the United States may not perfectly reflect US resource utilization. In this respect, however, because the primary cause of the cost differential is the eplerenone cost, cost calculations using the unit resource use of other countries should have little effect on the results. The ability to generalize the EPHESUS data to the wider population of patients with post-AMI heart failure is also uncertain. Finally, EPHESUS cannot be used to compare the clinical outcomes or cost-effectiveness of eplerenone to spironolactone. In EPHESUS, eplerenone was compared with placebo; there is no direct comparison of eplerenone to spironolactone, making a comparison of the 2 aldosterone blockers speculative.
The ICER in the subgroup of diabetic patients was higher than in other subgroups analyzed but still under the standard benchmark of $50 000. Diabetics were the only patients in EPHESUS who did not necessarily have to manifest evidence of heart failure to be enrolled in the trial. Approximately one third of EPHESUS subjects were diabetic, and about one third of this diabetic subgroup did not have evidence of heart failure. It is difficult to draw firm conclusions about the higher, but still acceptable, ICER in the relatively small diabetic subgroup compared with the overall trial population.
Eplerenone compares well to other therapies in terms of cost-effectiveness. The ICER for eplerenone is slightly higher than the ICER for clopidogrel in acute coronary syndromes24,25 but is similar to that of medications such as ACE inhibitors and ;-blockers. The ICER for captopril therapy versus no captopril in post-AMI patients 50 to 80 years of age is $3700 to $10 400, depending on age,26 whereas ;-blocker treatment after AMI has an ICER ranging from $360 to $17 000, depending on patient status.27 Note that in all studies of ;-blockers and ACE inhibitors after AMI, the patients were not on an aldosterone blocker, and concomitant standard therapies were different. Higher ICERs have also been noted with life-saving interventions; an ICER of $40 000 was noted in a comparison of an implantable cardiac defibrillator with amiodarone in survivors of cardiac arrest.28 With ICERs from $10 402 to $21 876 in cost per life-year gained, eplerenone compares well to other therapies in terms of cost-effectiveness and is below the $50 000 threshold used to determine whether society will consider the medication a good value.
Conclusions
The cost-effectiveness of eplerenone compares favorably to that of many other well-known and well-accepted therapies. Eplerenone is the only pharmacological agent proven to add incremental benefit on mortality and morbidity above and beyond standard therapy, including ACE inhibition and ;-blockers, in patients with heart failure after AMI. Despite some limitations, EPHESUS provides strong support for both the efficacy and cost-effectiveness of eplerenone compared with placebo in patients with post-MI heart failure. Eplerenone therapy should be considered a cost-effective component of the current armamentarium to improve survival in patients with heart failure after AMI.
Acknowledgments
This analysis was financially supported by Pfizer Inc, New York, NY.
References
American Heart Association. Heart Disease and Stroke Statistics—2004 Update. Dallas, Tex: American Heart Association; 2003.
Hasdai D, Topol EJ, Kilaru R, Battler A, Harrington RA, Vahanian A, Ohman EM, Granger CB, Van de Werf F, Simoons ML, O’Connor CM, Holmes DR Jr. Frequency, patient characteristics, and outcomes of mild-to-moderate heart failure complicating ST-segment elevation acute myocardial infarction: Lessons from 4 international fibrinolytic therapy trials. Am Heart J. 2003; 145: 73–79.
Steg PG, Dabbous OH, Feldman LJ, Steg PG, Dabbous OH, Feldman LJ, Cohen-Solal A, Aumont MC, Lopez-Sendon J, Budaj A, Goldberg RJ, Klein W, Anderson FA Jr, for the Global Registry of Acute Coronary Events Investigators. Determinants and prognostic impact of heart failure complicating acute coronary syndromes. Circulation. 2004; 109: 494–499.
Wu AH, Parsons L, Every NR, Bates ER, for the Second National Registry of Myocardial Infarction. Hospital outcomes in patients presenting with congestive heart failure complicating acute myocardial infarction: a report from the Second National Registry of Myocardial Infarction (NRMI-2). J Am Coll Cardiol. 2002; 40: 1389–1384.
Pitt B, Zannad F, Remme WJ, Cody R, Castaigne A, Perez A, Palensky J, Wittes J. The effect of spironolactone on morbidity and mortality in patients with severe heart failure: Randomized Aldactone Evaluation Study Investigators. N Engl J Med. 1999; 341: 709–717.
Pitt B, Remme W, Zannad F, Neaton J, Martinez F, Roniker B, Bittman R, Hurley S, Kleiman J, Gatlin M, for the Eplerenone Post-Acute Myocardial Infarction Heart Failure Efficacy and Survival Study Investigators. Eplerenone, a selective aldosterone blocker, in patients with left ventricular dysfunction after myocardial infarction. N Engl J Med. 2003; 348: 1309–1321.
Spertus JA, Tooley J, Jones P, Poston C, Mahoney E, Deedwania P, Hurley S, Pitt B, Weintraub WS. Expanding the outcomes in clinical trials of heart failure: the quality of life and economic components of EPHESUS (Eplerenone’s Neurohormonal Efficacy and Survival Study). Am Heart J. 2002; 143: 636–642.
Gold MR, Siegel JE, Russel LB, Weinstein MC. Cost-Effectiveness in Health and Medicine. New York, NY: Oxford University Press; 1996.
Drummond MF, O’Brien B, Stoddart GL, Torrance G. Methods for the Economic Evaluation of Health Care Programmes. Oxford, UK: Oxford Medical Publications; 1987.
Centers for Medicare and Medicaid Services. 100% MEDPAR Inpatient Hospital National Data for Fiscal Year 2001. Short Stay Inpatient by Diagnosis Related Groups. Available at: www.cms.gov/statistics/medpar/shtstay01drg.pdf. Accessed 2004.
Mitchell J, Burge RT, Lee AJ, McCall NT. Per Case Prospective Payment for Episodes of Hospital Care. Springfield, Va: US Department of Commerce National Technical Information Service; 1995.
Mahoney EM, Jurkovitz CT, Chu H, Becker ER, Culler S, Kosinski AS, Robertson DH, Alexander C, Nag S, Cook JR, Demopoulos LA, DiBattiste PM, Cannon CP, Weintraub WS, for the TACTICS-TIMI 18 Investigators. Treat Angina with Aggrastat and Determine Cost of Therapy With an Invasive or Conservative Strategy-Thrombolysis in Myocardial Infarction: cost and cost-effectiveness of an early invasive vs conservative strategy for the treatment of unstable angina and non-ST-segment elevation myocardial infarction. JAMA. 2002; 288: 1851–1858.
Rabin R, De Charro F. EQ-5D: a measure of health status from the EuroQoL group. Ann Med. 2001; 33: 337–343.
Weintraub WS, Mahoney EM, Zhang Z, Chu H, Hutton J, Buxton M, Booth J, Nugara F, Stables RH, Dooley P, Collinson J, Stuteville M, Delahunty N, Wright A, Flather MD. De Cock E. One year comparison of costs of coronary surgery versus percutaneous coronary intervention in the stent or surgery trial. Heart (Br Card Soc). 2004; 90: 782–788.
Peeters A, Mamun AA, Willekens F, Bonneux L. A cardiovascular life history: a life course analysis of the original Framingham Heart Study cohort. Eur Heart J. 2002; 23: 458–466.
Downey W, Beck P, McNutt M, Stang MR, Osei W, Nichol J. Health databases in Saskatchewan. In: Strom BL, ed. Pharmacoepidemiology. 3rd ed. Chichester, UK: Wiley; 2000; 325–345.
Goldberg RJ, Yarzebski J, Lessard D, Gore JM. A two-decades (1975 to 1995) long experience in the incidence, in-hospital and long-term case-fatality rates of acute myocardial infarction: a community-wide perspective. J Am Coll Cardiol. 1999; 33: 1533–1539.
Spencer FA, Meyer TE, Goldberg RJ, Yarzebski J, Hatton M, Lessard D, Gore JM. Twenty year trends (1975–1995) in the incidence, in-hospital and long-term death rates associated with heart failure complicating acute myocardial infarction: a community-wide perspective. J Am Coll Cardiol. 1999; 34: 1378–1387.
Blackstone EH, Naftel DC, Turner MEJ. The decomposition of time-varying hazard into phases, each incorporating a separate stream of concomitant information. J Am Stat Assn. 1986; 81: 615–624.
Meltzer D. Accounting for future costs in medical cost-effectiveness analysis. J Health Econ. 1997; 16: 33–64.
Efron B, Tibshirani R. An Introduction to the Bootstrap. New York, NY: Chapman and Hall; 1993.
Mahoney EM, Chu H. Cost-Effectiveness analysis alongside clinical trials: statistical and methodologic issues. In: Weintraub WS, ed. Cardiovascular Health Care Economics. Totowa, NJ: Humana Press, Inc.; 2003: 123–156.
Weintraub WS, Cole J, Tooley JF. Cost and cost-effectiveness studies in heart failure research. Am Heart J. 2002; 143: 565–576.
Latour-Perez J, Navarro-Ruiz A, Ridao-Lopez M, Cervera-Montes M. Using clopidogrel in non-ST-segment elevation acute coronary syndrome patients: a cost-utility analysis in Spain. Value Health. 2004; 7: 52–60.
Lindgren P, Jonsson B, Yusuf S. Cost-effectiveness of clopidogrel in acute coronary syndromes in Sweden: a long-term model based on the cure trial. J Intern Med. 2004; 255: 562–570.
Tsevat J, Duke D, Goldman L, Pfeffer MA, Lamas GA, Soukup JR, Kuntz KM, Lee TH. Cost-effectiveness of captopril therapy after myocardial infarction. J Am Coll Cardiol. 1995; 26: 914–919.
Tengs TO, Adams ME, Pliskin JS, Safran DG, Siegel JE, Weinstein MC, Graham JD. Five-hundred life-saving interventions and their cost-effectiveness. Risk Anal. 1995; 15: 369–390.
Owens DK, Sanders GD, Harris RA, McDonald KM, Heidenreich PA, Dembitzer AD, Hlatky MA. Cost-effectiveness of implantable cardioverter defibrillators relative to amiodarone for prevention of sudden cardiac death. Ann Intern Med. 1997; 126: 1–12.(William S. Weintraub, MD;)
New England Research Institutes, Watertown, Mass (E.M.M.)
Mid America Heart Institute, Kansas City, Mo (J.A.S.)
Caro Research, Inc., Boston, Mass (J.C., J.I.)
University of Massachusetts, Worcester (R.G.)
Prudential Equity Group, Deerfield, Ill (J.T.)
Pfizer, Inc, New York, NY (R.W.)
University of Michigan, Ann Arbor (B.P.).
Abstract
Background— In the Eplerenone Post-Acute Myocardial Infarction Heart Failure Efficacy and Survival Study (EPHESUS), aldosterone blockade with eplerenone decreased mortality in patients with left ventricular systolic dysfunction and heart failure after acute myocardial infarction. The present study was performed to evaluate the cost-effectiveness of eplerenone compared with placebo in these patients.
Methods and Results— A total of 6632 patients with left ventricular systolic dysfunction and heart failure after acute myocardial infarction were randomized to eplerenone or placebo and followed up for a mean of 16 months. The coprimary end points were all-cause mortality and the composite of cardiovascular mortality/cardiovascular hospitalization. The evaluation of resource use included hospitalizations, outpatient services, and medications. Eplerenone was priced at the average wholesale price, $3.60 per day. Survival beyond the trial period was estimated from data from the Framingham Heart Study, the Saskatchewan Health database, and the Worcester Heart Attack Registry. The incremental cost-effectiveness of eplerenone in cost per life-year and quality-adjusted life-year gained compared with placebo was estimated. The number of life-years gained with eplerenone was 0.1014 based on Framingham (95% CI, 0.0306 to 0.1740), 0.0636 with Saskatchewan (95% CI, 0.0229 to 0.1038), and 0.1337 with Worcester (95% CI, 0.0438 to 0.2252) data. Cost was $1391 higher over the trial period in the eplerenone arm (95% CI, 656 to 2165) because of drug cost. The incremental cost-effectiveness ratio was $13 718 per life-year gained with Framingham (96.7% under $50 000 per life-year gained), $21 876 with Saskatchewan, and $10 402 with Worcester.
Conclusions— Eplerenone compared with placebo in the treatment of heart failure after acute myocardial infarction is effective in reducing mortality and is cost-effective in increasing years of life by commonly used criteria.
Key Words: cost-benefit analysis ; heart failure ; myocardial infarction
Introduction
One of the most serious and frequent consequences of acute myocardial infarction (AMI) is heart failure, which develops in 22% of men and 46% of women after an MI.1 The presence of heart failure in patients with AMI is associated with a 55% greater risk of dying and 2.15-times-greater risk of death or recurrent AMI at 30 days.2 Patients with AMI who present to the hospital with heart failure have longer hospital stays, higher readmission rates, and higher mortality rates during hospitalization and 6 months after discharge than those without heart failure.3,4 With estimated direct and indirect health expenditures for heart failure nearing $26 billion annually,1 cost-effective treatment strategies for this disease are needed.
Multiple therapeutic strategies have been used to prolong life and to decrease hospitalizations for heart failure, including diuretics, ACE inhibitors or angiotensin receptor blockers (ARBs), ;-blockers, resynchronization therapy, implantable cardiac defibrillators, and heart transplantation. Additionally, nonselective aldosterone blockade has been shown to reduce mortality in patients with chronic, severe heart failure when used with ACE inhibitors, diuretics, and sometimes digoxin.5 Recently, the Eplerenone Post-Acute Myocardial Infarction Heart Failure Efficacy and Survival Study (EPHESUS)6 demonstrated that selective aldosterone blockade with eplerenone significantly reduced mortality and morbidity in patients with left ventricular systolic dysfunction (LVSD) and heart failure after AMI who were receiving optimal medical therapies. This trial was the first to demonstrate incremental benefit of a therapeutic agent in addition to standard therapy (including an ACE inhibitor or ARB and a ;-blocker) for improving outcomes in patients with heart failure after AMI.
Although its clinical efficacy is clear, the question remains as to whether the added benefit of eplerenone is worth the added cost. This study presents a cost-effectiveness analysis to define the value of the use of eplerenone compared with placebo in patients with heart failure after an AMI.
Methods
EPHESUS Trial Design
EPHESUS, a double-blind, multicenter, randomized, placebo-controlled trial, has previously been described in detail.6 Briefly, 6632 patients were recruited from December 1999 to December 2001 at 671 centers in 37 countries. Patients were randomized to eplerenone or placebo from 3 to 14 days after AMI. Inclusion criteria included LVSD (documented ejection fraction 40%) and heart failure documented by pulmonary rales, venous congestion on chest x-ray, or the presence of a third heart sound. Patients with diabetes were required to have LVSD but were not required to have documentation of heart failure. Patients were randomized to eplerenone (25 mg/d) or placebo for 4 weeks, after which the dosage of eplerenone was increased to a maximum of 50 mg/d. Patients received standard optimal medical therapy, which could include ACE inhibitors or ARBs, diuretics, ;-blockers, statin therapy, and coronary reperfusion. The 2 primary end points were time to death resulting from any cause and time to death caused by cardiovascular causes or first hospitalization for a cardiovascular event, including heart failure, recurrent AMI, stroke, or ventricular arrhythmia. The major secondary end points included death resulting from cardiovascular causes and death resulting from any cause or any hospitalization.
Economic Analysis and Costs
The economic analytic plan of EPHESUS was to compare the costs of the 2 treatment arms and, if the eplerenone arm was more costly and more effective, to perform an incremental cost-effectiveness analysis.7 Although not all sources of costs could be accounted for, the overall perspective was societal. Costs included in the analysis were direct medical care costs for hospitalizations, outpatient procedures, and drugs.8,9 No data were available from EPHESUS that could be used to calculate indirect costs resulting from lost productivity. All costs used 2001 as the base year, except the cost of eplerenone, which was not marketed until 2004. Costs and life expectancy differences were discounted 3% annually. The analysis used American unit costs but used resource use information and clinical outcomes for all 6632 patients. Cardiovascular healthcare resource use associated with the index and all follow-up hospitalizations, outpatient diagnostic tests and procedures, and medications were recorded prospectively.
Using a predefined algorithm designed by the investigators from case report form resource use, an investigator blinded to treatment group assigned the initial and subsequent hospitalizations for patients enrolled in EPHESUS to a diagnosis-related group (DRG) as used in the Medicare program in the United States. Costs for each DRG were estimated from average Medicare reimbursement rates obtained from the Medicare Part A data file,10 and professional costs were calculated by percent share by DRG according to the method of Mitchell et al.11 An investigator blinded to treatment group coded outpatient procedures by current procedural terminology and assigned a cost based on the Medicare fee schedule. All medications were assigned a cost based on Redbook average wholesale price (AWP).12 Specifically, the AWP for eplerenone was $3.60. All medications were assumed to continue for the duration of time that each patient was followed up.
Utility was measured with the EQ-5D13,14 in a subset from English-speaking countries of 1792 patients at baseline, 1530 patients at 6 months, and 1123 patients at 12 months. Quality-adjusted life-years (QALYs) were then calculated by multiplying survival by utility. For patients with a missing utility score, the average utility for all patients with available scores by treatment arm was used to estimate utility. Utility after 12 months was carried forward by use of the 12-month value. Because utility was measured in only a minority of patients, the primary analysis of cost-effectiveness remained cost per life-year gained, with cost per QALY gained as a sensitivity analysis.
Lifetime cost-effectiveness ratios in terms of cost per life-year gained and cost per QALY gained were predicted from in-trial estimates of incremental costs, event rates (death), and estimates of lost life expectancy associated with those in-trial deaths obtained from 3 sources: the Framingham Heart Study,12,15 the Saskatchewan Health database,16 and the Worcester Heart Attack Registry.17,18 These 3 sources were used to estimate survival because no single source perfectly met these criteria. For the Saskatchewan and Worcester databases, data on 2543 and 1094 patients, respectively, with heart failure after an AMI were analyzed with fractional polynomials and piecewise regression to obtain death hazard functions over time.19 These functions were adjusted according to patient characteristics through the use of separate Cox proportional-hazards models derived from the same data. For patients who died during the trial, life-years lost were obtained by subtracting the in-trial survival times from estimated age- and sex-specific life expectancy estimates.12 Patients were considered to have 0 life-years lost if they survived during the trial period. Average life-years lost for each treatment group were calculated across all patients who died and survived in each arm of the trial. The difference in average life-years lost because of deaths (placebo minus eplerenone) yields an estimate of the life-years gained with eplerenone. Life-years and QALYs were discounted at 3% annually.12 The additional healthcare costs attributed to life-years gained by treatment were estimated as a sensitivity analysis.20 Costs beyond the trial period were estimated by calculating the costs during the trial by year, carrying forward the average cost in years 2 and 3 of the trial, and discounting by 3% annually. Bootstrap methods were used to estimate the fraction of the joint distribution of the cost and effectiveness differences lying in different regions of the cost-effectiveness plane.21,22
In addition to applying these methods to the overall population, we also performed these cost-effectiveness analyses for certain demographic subgroups defined by age, sex, diabetes, and prior AMI.
Results
There were no differences in the baseline characteristics of age, gender, prior AMI, diabetes, hypertension, prior history of heart failure, and ejection fraction (Table 1). There was a 15% relative decrease in death from any cause with eplerenone compared with placebo. There was a 13% relative decrease in the other primary end point of death or hospitalization for cardiovascular events and significant decreases in the secondary end points of death from any cause or any hospitalization, sudden death from cardiac causes, and the number of episodes of hospitalization for heart failure. The average follow-up was 16 months.
The incremental cost-effectiveness ratio (ICER) of eplerenone compared with placebo in the analysis with Framingham was $13 718 per life-year gained, with 96.7% of estimates falling below the threshold of $50 000 per life-year gained (Table 6). When the Saskatchewan estimates of life expectancy were used, the ICER was $21 876, with 93.8% of estimates under $50 000 per life-year gained. Based on Worcester estimates of life expectancy, the ICER was $10 402 per life-year gained, with 98.8% of observations falling below the threshold of $50 000 per life-year gained. The ICERs were systematically higher when calculated in cost per QALY gained. The ICERs were also systematically higher if the costs after the trial period were included. The joint bootstrap distribution of the difference in efficacy in life-years and cost is displayed in Figure 1, with the Framingham estimates for lost life expectancy used. Almost all estimates are in quadrant 1 of the cost-effectiveness plane, meaning that there was greater efficacy at increased cost with eplerenone. The diagonal line from the origin represents $50 000 per life-year gained. Estimates below this line in quadrant 1 or 2 would be cost-effective, if $50 000 represents society’s willingness-to-pay threshold.
Discussion
This analysis revealed that eplerenone in the setting of heart failure after an acute MI is a life-saving medication that is cost-effective compared with placebo by the common benchmark ceiling ratio of $50 000 per life-year gained. This conclusion was robust throughout a range of projections using 3 different sources for estimates of lost life expectancy resulting from in-trial deaths in a patient population in which the majority received both ;-blockers (75%) and ACE inhibitors or ARBs (87%).6 Furthermore, the bootstrap analysis revealed that with each method of costing, >90% of simulations were below the $50 000 benchmark.
Previous reports from EPHESUS documented improved survival and fewer cardiovascular events in patients experiencing heart failure after an AMI who are treated with eplerenone compared with placebo.6 The results from EPHESUS complement previous results from the Randomized ALdactone (spironolactone) Evaluation Study for congestive heart failure (RALES), which showed improved survival in patients with severe chronic heart failure who were treated with spironolactone and in whom recent MI was excluded.5 EPHESUS is the first and only study to demonstrate the efficacy of aldosterone blockade for reducing mortality and morbidity in post-AMI patients with heart failure. It is important to note that eplerenone was effective in preventing events in these patients who were already treated optimally with ;-blockers and either ACE inhibitors or ARBs and that this is the only agent proven to add incremental benefit on mortality and morbidity above and beyond standard therapy in these patients;6 ;-blockers and ACE inhibitors have been shown to be clinically effective and cost-effective in the treatment of heart failure.23 The present study reveals that compared with placebo, eplerenone is cost-effective in the treatment of post-MI heart failure in a population of patients already receiving standard therapy.
Study Limitations
The follow-up period in EPHESUS was of variable length (range, 0 to 33 months). Thus, cost-effectiveness was calculated for the average follow-up period of 16 months, with therapy with eplerenone for that period of time but not thereafter. This study does not and realistically cannot address the issue of how long eplerenone should be taken. The estimation of life expectancy assumes that the survival curves remain parallel after the trial period.
The in-trial results cannot give a full picture of the survival advantage of eplerenone. Thus, the estimation of survival was extended beyond the study period by using data from 3 separate sources. However, the degree to which the survival experience of patients from these observational studies yields accurate estimates of life expectancy for the EPHESUS population is uncertain. In both the Saskatchewan and Worcester databases, mortality within 1 year of MI was higher than in EPHESUS. This would mean a shorter projected life expectancy, rendering our results conservative. In addition, there was considerable variation between the estimates from these 3 sources, given variations in the populations. The Framingham Heart Study was used because it is a large, well-known, epidemiological database.15 The Worcester and Saskatchewan databases also were chosen because they included patients similar to those in EPHESUS (ie, post-AMI heart failure patients), and both databases included long-term data.16–18 Although the Worcester database had the greatest similarity to EPHESUS and the closest estimate of mortality hazard, the ICERs from these 3 databases were all in an acceptable range and provide a sensitivity analysis supporting the cost-effectiveness of eplerenone.
The costing methodology was based on Medicare payments for hospitalization, the Medicare fee schedule for procedures, and AWPs for medications. Eplerenone was priced at the AWP of $3.60 a day. The extent to which these costs reflect resource use from a societal point of view is somewhat uncertain because there is no single source for costs that unequivocally represents societal costs. Thus, Medicare payments may appropriately reflect costs for the Centers for Medicare and Medicaid Services but may not adequately represent resource consumption by hospitals and physicians because Medicare costs tend to be lower than managed care costs. Therefore, the present analysis may provide a conservative representation of the cost-effectiveness of eplerenone.
Costing beyond the trial period was based on projections of costs within the trial period. This approach, while reasonable, is not easily subject to empirical assessment.
Indirect costs were not included, especially those reflecting return to work. To the extent that eplerenone permits return to work by preventing hospitalizations or deaths, this would lower costs in the eplerenone arm, again rendering the results conservative.
Resource use and clinical outcomes were considered trial wide. Use of Medicare costing does not account for possible within-DRG differences in treatment practices and resource use across countries. With so many countries, it is not possible to adequately account for variation in DRGs across countries. Thus, the use of trial-wide data for costing in the United States may not perfectly reflect US resource utilization. In this respect, however, because the primary cause of the cost differential is the eplerenone cost, cost calculations using the unit resource use of other countries should have little effect on the results. The ability to generalize the EPHESUS data to the wider population of patients with post-AMI heart failure is also uncertain. Finally, EPHESUS cannot be used to compare the clinical outcomes or cost-effectiveness of eplerenone to spironolactone. In EPHESUS, eplerenone was compared with placebo; there is no direct comparison of eplerenone to spironolactone, making a comparison of the 2 aldosterone blockers speculative.
The ICER in the subgroup of diabetic patients was higher than in other subgroups analyzed but still under the standard benchmark of $50 000. Diabetics were the only patients in EPHESUS who did not necessarily have to manifest evidence of heart failure to be enrolled in the trial. Approximately one third of EPHESUS subjects were diabetic, and about one third of this diabetic subgroup did not have evidence of heart failure. It is difficult to draw firm conclusions about the higher, but still acceptable, ICER in the relatively small diabetic subgroup compared with the overall trial population.
Eplerenone compares well to other therapies in terms of cost-effectiveness. The ICER for eplerenone is slightly higher than the ICER for clopidogrel in acute coronary syndromes24,25 but is similar to that of medications such as ACE inhibitors and ;-blockers. The ICER for captopril therapy versus no captopril in post-AMI patients 50 to 80 years of age is $3700 to $10 400, depending on age,26 whereas ;-blocker treatment after AMI has an ICER ranging from $360 to $17 000, depending on patient status.27 Note that in all studies of ;-blockers and ACE inhibitors after AMI, the patients were not on an aldosterone blocker, and concomitant standard therapies were different. Higher ICERs have also been noted with life-saving interventions; an ICER of $40 000 was noted in a comparison of an implantable cardiac defibrillator with amiodarone in survivors of cardiac arrest.28 With ICERs from $10 402 to $21 876 in cost per life-year gained, eplerenone compares well to other therapies in terms of cost-effectiveness and is below the $50 000 threshold used to determine whether society will consider the medication a good value.
Conclusions
The cost-effectiveness of eplerenone compares favorably to that of many other well-known and well-accepted therapies. Eplerenone is the only pharmacological agent proven to add incremental benefit on mortality and morbidity above and beyond standard therapy, including ACE inhibition and ;-blockers, in patients with heart failure after AMI. Despite some limitations, EPHESUS provides strong support for both the efficacy and cost-effectiveness of eplerenone compared with placebo in patients with post-MI heart failure. Eplerenone therapy should be considered a cost-effective component of the current armamentarium to improve survival in patients with heart failure after AMI.
Acknowledgments
This analysis was financially supported by Pfizer Inc, New York, NY.
References
American Heart Association. Heart Disease and Stroke Statistics—2004 Update. Dallas, Tex: American Heart Association; 2003.
Hasdai D, Topol EJ, Kilaru R, Battler A, Harrington RA, Vahanian A, Ohman EM, Granger CB, Van de Werf F, Simoons ML, O’Connor CM, Holmes DR Jr. Frequency, patient characteristics, and outcomes of mild-to-moderate heart failure complicating ST-segment elevation acute myocardial infarction: Lessons from 4 international fibrinolytic therapy trials. Am Heart J. 2003; 145: 73–79.
Steg PG, Dabbous OH, Feldman LJ, Steg PG, Dabbous OH, Feldman LJ, Cohen-Solal A, Aumont MC, Lopez-Sendon J, Budaj A, Goldberg RJ, Klein W, Anderson FA Jr, for the Global Registry of Acute Coronary Events Investigators. Determinants and prognostic impact of heart failure complicating acute coronary syndromes. Circulation. 2004; 109: 494–499.
Wu AH, Parsons L, Every NR, Bates ER, for the Second National Registry of Myocardial Infarction. Hospital outcomes in patients presenting with congestive heart failure complicating acute myocardial infarction: a report from the Second National Registry of Myocardial Infarction (NRMI-2). J Am Coll Cardiol. 2002; 40: 1389–1384.
Pitt B, Zannad F, Remme WJ, Cody R, Castaigne A, Perez A, Palensky J, Wittes J. The effect of spironolactone on morbidity and mortality in patients with severe heart failure: Randomized Aldactone Evaluation Study Investigators. N Engl J Med. 1999; 341: 709–717.
Pitt B, Remme W, Zannad F, Neaton J, Martinez F, Roniker B, Bittman R, Hurley S, Kleiman J, Gatlin M, for the Eplerenone Post-Acute Myocardial Infarction Heart Failure Efficacy and Survival Study Investigators. Eplerenone, a selective aldosterone blocker, in patients with left ventricular dysfunction after myocardial infarction. N Engl J Med. 2003; 348: 1309–1321.
Spertus JA, Tooley J, Jones P, Poston C, Mahoney E, Deedwania P, Hurley S, Pitt B, Weintraub WS. Expanding the outcomes in clinical trials of heart failure: the quality of life and economic components of EPHESUS (Eplerenone’s Neurohormonal Efficacy and Survival Study). Am Heart J. 2002; 143: 636–642.
Gold MR, Siegel JE, Russel LB, Weinstein MC. Cost-Effectiveness in Health and Medicine. New York, NY: Oxford University Press; 1996.
Drummond MF, O’Brien B, Stoddart GL, Torrance G. Methods for the Economic Evaluation of Health Care Programmes. Oxford, UK: Oxford Medical Publications; 1987.
Centers for Medicare and Medicaid Services. 100% MEDPAR Inpatient Hospital National Data for Fiscal Year 2001. Short Stay Inpatient by Diagnosis Related Groups. Available at: www.cms.gov/statistics/medpar/shtstay01drg.pdf. Accessed 2004.
Mitchell J, Burge RT, Lee AJ, McCall NT. Per Case Prospective Payment for Episodes of Hospital Care. Springfield, Va: US Department of Commerce National Technical Information Service; 1995.
Mahoney EM, Jurkovitz CT, Chu H, Becker ER, Culler S, Kosinski AS, Robertson DH, Alexander C, Nag S, Cook JR, Demopoulos LA, DiBattiste PM, Cannon CP, Weintraub WS, for the TACTICS-TIMI 18 Investigators. Treat Angina with Aggrastat and Determine Cost of Therapy With an Invasive or Conservative Strategy-Thrombolysis in Myocardial Infarction: cost and cost-effectiveness of an early invasive vs conservative strategy for the treatment of unstable angina and non-ST-segment elevation myocardial infarction. JAMA. 2002; 288: 1851–1858.
Rabin R, De Charro F. EQ-5D: a measure of health status from the EuroQoL group. Ann Med. 2001; 33: 337–343.
Weintraub WS, Mahoney EM, Zhang Z, Chu H, Hutton J, Buxton M, Booth J, Nugara F, Stables RH, Dooley P, Collinson J, Stuteville M, Delahunty N, Wright A, Flather MD. De Cock E. One year comparison of costs of coronary surgery versus percutaneous coronary intervention in the stent or surgery trial. Heart (Br Card Soc). 2004; 90: 782–788.
Peeters A, Mamun AA, Willekens F, Bonneux L. A cardiovascular life history: a life course analysis of the original Framingham Heart Study cohort. Eur Heart J. 2002; 23: 458–466.
Downey W, Beck P, McNutt M, Stang MR, Osei W, Nichol J. Health databases in Saskatchewan. In: Strom BL, ed. Pharmacoepidemiology. 3rd ed. Chichester, UK: Wiley; 2000; 325–345.
Goldberg RJ, Yarzebski J, Lessard D, Gore JM. A two-decades (1975 to 1995) long experience in the incidence, in-hospital and long-term case-fatality rates of acute myocardial infarction: a community-wide perspective. J Am Coll Cardiol. 1999; 33: 1533–1539.
Spencer FA, Meyer TE, Goldberg RJ, Yarzebski J, Hatton M, Lessard D, Gore JM. Twenty year trends (1975–1995) in the incidence, in-hospital and long-term death rates associated with heart failure complicating acute myocardial infarction: a community-wide perspective. J Am Coll Cardiol. 1999; 34: 1378–1387.
Blackstone EH, Naftel DC, Turner MEJ. The decomposition of time-varying hazard into phases, each incorporating a separate stream of concomitant information. J Am Stat Assn. 1986; 81: 615–624.
Meltzer D. Accounting for future costs in medical cost-effectiveness analysis. J Health Econ. 1997; 16: 33–64.
Efron B, Tibshirani R. An Introduction to the Bootstrap. New York, NY: Chapman and Hall; 1993.
Mahoney EM, Chu H. Cost-Effectiveness analysis alongside clinical trials: statistical and methodologic issues. In: Weintraub WS, ed. Cardiovascular Health Care Economics. Totowa, NJ: Humana Press, Inc.; 2003: 123–156.
Weintraub WS, Cole J, Tooley JF. Cost and cost-effectiveness studies in heart failure research. Am Heart J. 2002; 143: 565–576.
Latour-Perez J, Navarro-Ruiz A, Ridao-Lopez M, Cervera-Montes M. Using clopidogrel in non-ST-segment elevation acute coronary syndrome patients: a cost-utility analysis in Spain. Value Health. 2004; 7: 52–60.
Lindgren P, Jonsson B, Yusuf S. Cost-effectiveness of clopidogrel in acute coronary syndromes in Sweden: a long-term model based on the cure trial. J Intern Med. 2004; 255: 562–570.
Tsevat J, Duke D, Goldman L, Pfeffer MA, Lamas GA, Soukup JR, Kuntz KM, Lee TH. Cost-effectiveness of captopril therapy after myocardial infarction. J Am Coll Cardiol. 1995; 26: 914–919.
Tengs TO, Adams ME, Pliskin JS, Safran DG, Siegel JE, Weinstein MC, Graham JD. Five-hundred life-saving interventions and their cost-effectiveness. Risk Anal. 1995; 15: 369–390.
Owens DK, Sanders GD, Harris RA, McDonald KM, Heidenreich PA, Dembitzer AD, Hlatky MA. Cost-effectiveness of implantable cardioverter defibrillators relative to amiodarone for prevention of sudden cardiac death. Ann Intern Med. 1997; 126: 1–12.(William S. Weintraub, MD;)