Skip Navigation

European Heart Journal 2006 27(12):1447-1458; doi:10.1093/eurheartj/ehl016
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow E-letters: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when E-letters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (6)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by McMurray, J. J.V.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by McMurray, J. J.V.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The European Society of Cardiology 2006. All rights reserved. For Permissions, please e-mail: journals.permissions@oxfordjournals.org

Resource utilization and costs in the Candesartan in Heart failure: Assessment of Reduction in Mortality and morbidity (CHARM) programme

John J.V. McMurray1,*, Fredrik L. Andersson2, Simon Stewart1, Klas Svensson2, Alain Cohen Solal3, Rainer Dietz4, Johan Vanhaecke5, Dirk J. van Veldhuisen6, Jan Östergren7, Christopher B. Granger8, Salim Yusuf9, Marc A. Pfeffer10, Karl Swedberg11 for the CHARM Investigators and Committees

1 Department of Cardiology, Level 4, Western Infirmary, Glasgow G11 6NT, Scotland, UK
2 AstraZeneca R&D, Lund, Sweden
3 Deparment of Cardiology, Hôpital Beaujon, Clichy, France
4 Franz-Volhard-Klinik, Berlin, Germany
5 University Hospital Gasthuisberg, Belgium
6 Department of Cardiology, University Hospital Groningen, Groningen, The Netherlands
7 Department of Medicine, Karolinska University Hospital, Stockholm, Sweden
8 Duke University Medical Center, Durham, NC USA
9 HGM-McMaster Clinic, Hamilton, Ontario, Canada
10 Cardiovascular Division, Brigham and Women's Hospital, Boston, MA USA
11 Department of Medicine, Sahlgrenska University Hospital, Göteborg, Sweden

Received 29 September 2005; revised 8 March 2006; accepted 13 April 2006.

* Corresponding author. Tel: +44 141 211 1838; fax: +44 141 211 2252. E-mail address: j.mcmurray{at}bio.gla.ac.uk


   Abstract
Top
 Abstract
Introduction
 Methods
 Results
 Discussion
 Acknowledgements
 References
 
Aims More treatments are needed to improve clinical outcomes in chronic heart failure (HF). It is, however, important that treatments for a condition as common as HF are affordable. We have carried out a prospective economic analysis of the Candesartan in Heart failure: Assessment of Reduction in Mortality and morbidity (CHARM) programme.

Methods and results Patients with NYHA class II–IV HF and LVEF ≤0.40 were randomized to CHARM-Alternative if intolerant of an ACE-inhibitor or to CHARM-Added if taking an ACE-inhibitor. Patients with a LVEF >0.40 were randomized in CHARM-Preserved. Each trial compared the effect of candesartan to placebo on the primary outcome of cardiovascular death or HF hospitalization. Detailed information was prospectively collected on hospital admissions, procedures/operations and drugs. A cost–consequence analysis was performed for France, Germany and the UK for CHARM-Overall and a cost-effectiveness analysis for the low LVEF trials. The cost of candesartan was substantially offset by a reduction in hospital admissions, especially for HF. In the cost–consequence analysis, candesartan was cost-saving in most scenarios for CHARM-Alternative and Added but the marginal annual net cost per patient was upto {euro}372 per year in CHARM-Preserved, in which candesartan did not reduce the primary outcome significantly. In the cost-effectiveness analysis of patients with a LVEF ≤ 0.40, candesartan was cost-saving in some scenarios and in the others the maximum cost per life year gained was {euro}3881.

Conclusion Candesartan improves functional class, reduces the risk of hospital admission, and increases survival in patients with a HF and a LVEF ≤0.40 at an acceptable cost.

Key Words: Heart failure • Angiotensin receptor blocker • Health economics • Cost-effectiveness


   Introduction
Top
 Abstract
 Introduction
Methods
 Results
 Discussion
 Acknowledgements
 References
 
Despite the availability of several effective treatments,1,2 patients with chronic heart failure (HF) continue to experience marked functional limitation because of symptoms, reduction in quality of life, frequent admission to hospital, and greatly shortened life-expectancy.3–7 Consequently, HF continues to be a major public health problem, for which new treatments are needed.8 For a condition as prevalent as HF, the cost of any new treatment (and whether it is justified) is, inevitably questioned.9–14 We have recently shown that the angiotensin II type 1 receptor blocker (ARB) candesartan reduces symptomatic limitation and decreases the risk of hospital admission for worsening HF and deaths due to cardiovascular causes, when added to conventional treatment in a broad spectrum of patients with symptomatic HF randomized in the Candesartan in Heart failure: Assessment of Reduction in Mortality and morbidity (CHARM) programme.15–21 An economic evaluation of this programme was pre-specified and data on resource use were collected prospectively. Here we describe within-trial cost–consequence and cost-effectiveness analyses of CHARM.15


   Methods
Top
 Abstract
 Introduction
 Methods
Results
 Discussion
 Acknowledgements
 References
 
The CHARM programme
The design, baseline findings, and primary results of the CHARM programme have been reported in detail.15–21 Briefly, the CHARM programme consisted of three independent but related trials in which patients with NYHA class II–IV HF were randomized to placebo or candesartan (target dose 32 mg once daily). In CHARM-Added, patients in NYHA Class II had to have had a hospital admission for a cardiac reason in the previous 6 months (this had the effect of increasing the proportion of NYHA class III/IV patients in CHARM-Added). Patients were enrolled into the individual CHARM trials according to left ventricular (LV) ejection fraction (EF) (LVEF) and baseline treatment with an ACE-inhibitor. Patients with an LVEF ≤0.40 and intolerant of an ACE-inhibitor, were enrolled in CHARM-Alternative,17 whereas patients with an LVEF ≤0.40 and taking an ACE-inhibitor were enrolled in CHARM-Added.18 Patients with an LVEF >0.40 were randomized into CHARM-Preserved.19 The CHARM programme was completed, as planned, 2 years after the last patient was randomized. Because the rate of recruitment varied between the CHARM trials, overall follow-up ranged from a median of 41 [interquartile (IQR) range 35–44] months in CHARM-Added, to 37 (34–41) months in CHARM-Preserved, 34 (27–41) months in CHARM-Alternative, and 38 (32–42) months in the overall CHARM programme.17–20

Economic analysis of CHARM
These analyses were based on the comparison of placebo, i.e. conventional treatment for HF (combination of a diuretic, digoxin, ACE-inhibitor, beta-blocker, and spironolactone, as indicated and tolerated) to candesartan added to conventional treatment. We examined the effect of adding candesartan in all 7599 patients randomized in 26 countries. For the purposes of the economic analysis, however, all patients were considered to have been managed in France, Germany, or the UK. Our analysis takes the perspective of a third party payer in France and Germany and the National Health Service (NHS) in the UK. Information was collected prospectively on drug treatment, patients admitted to hospital (proportion admitted, number of admissions per patient, number of hospital days per patient), admissions for cardiovascular reasons (number, duration, ward type), and procedures/operations as described below. These data were used to determine the additional direct costs incurred, and potential savings made with candesartan. Indirect costs such as loss of productivity and earnings due to inability to work were not considered. The analysis period was that of the trial concerned and the programme overall; no future projections were made.

Effectiveness measures
As described in detail elsewhere, the primary clinical outcome in the component trials of the CHARM programme was the composite of cardiovascular death or hospital admission for worsening HF.15–21 The candesartan to placebo hazard ratio for this outcome in CHARM-Alternative was 0.77 (95% CI 0.67–0.89, P=0.0004),17 in CHARM-Added 0.85 (0.75–0.96, P=0.011),18 in CHARM-Preserved 0.89 (0.77–1.03, P=0.118),19 and in the overall CHARM programme 0.84 (0.77–0.91, P<0.0001).20

All-cause mortality was the pre-specified primary endpoint of the overall CHARM programme and of the two reduced LVEF trials combined (CHARM-Alternative and Added).15 The candesartan to placebo hazard ratio for this outcome in the overall CHARM programme was 0.91 (0.83–1.00,20 P=0.055) and in the reduced LVEF trials 0.88 (0.79–0.98, P=0.018).21

Resource use
Hospital admissions
For cardiovascular admissions, investigators were asked to specify the type of ward (intensive/coronary care unit, cardiology, general internal medicine) patients were admitted to and the number of days spent in each type of ward.

For non-cardiovascular admissions only, length of stay (and not ward type) was collected. For these admissions, we assumed that 10% of time was spent in intensive care and 90% on an internal medicine ward.

Procedures
Investigators were asked to provide information on cardiovascular procedures and operations (a checklist menu was provided). Non-cardiovascular procedures were not recorded and were assumed to be equal in the two groups.

Drug treatment
Investigators classified medication as cardiovascular or non-cardiovascular. We assumed that non-cardiovascular medication did not differ between treatment groups (and was not, therefore, used in this analysis). ‘Cardiovascular’ medication included digitalis glycosides, diuretics, beta-blockers, calcium channel blockers, other vasodilators, anti-arrhythmic drugs, ACE-inhibitors, ARBs, and other cardiovascular drugs such as lipid-lowering agents and anticoagulants.

Estimation of costs
Hospital admissions
Two approaches were used, as described previously (Table 1).22

  • Diagnosis-related group (DRG) costs were obtained from government sources in France, Germany, and the UK.23–28
  • Per diem (hospital bed-day) costs were also obtained from official sources in these countries.23–28 In this analysis actual recorded days in hospital were multiplied by the daily unit costs of hospital care.


View this table:
[in this window]
[in a new window]
  		Table 1 Unit costs used in the economic analysis of CHARM

 
Drug treatment
Drug costs were obtained from standard tariffs for the financial year 2003–04 (Vidal, France; Rote Liste, Germany; British National Formulary, UK).29–32 Daily dose was multiplied by the cost per dose to calculate the daily cost of treatment. Where available, the costs of generic drugs were used. Dispensing fees were included in the public price of candesartan. We added costs to account for the initiation and up-titration of candesartan, namely four extra GP visits and four checks of blood biochemistry, as previously described.22

Cardiovascular procedures
The DRG costs of cardiovascular procedures and operations were obtained from local government sources.23–29 Because little reliable and comparable public information is available for the per diem costs of cardiovascular procedures, DRG costs were used as proxies in the per diem analysis.

All costs were converted to 2003 values using the local price index with the exception of drug prices which are for 2003–04. Costs are presented in the local currency (£ and {euro}) where £1=1.79 US$ and 1{euro}=1.20 US$. The recommended discount rate of 3% was used.33

The costs used in this analysis are summarized in Table 1.

Economic analyses performed
Two types of economic analysis were performed:

  • A cost–consequence analysis was performed for each component trial and for the overall CHARM programme. In this, we used the primary outcome of the component trials as the measure of effectiveness i.e. we calculated the cost per patient treated to postpone or prevent one patient experiencing a cardiovascular death or hospital admission for worsening HF within the trial.34
  • Incremental cost-effectiveness ratios [cost per life year gained (LYG)] were calculated for the reduced LVEF trials as, in these two trials combined, there was a significant increase in survival with candesartan. This analysis was not performed for CHARM-Preserved as there was no reduction in cardiovascular or all cause mortality in that trial with candesartan.

Sensitivity analyses
The following sensitivity analyses were conducted: (i) we increased the length of non-cardiovascular hospital stay by 30% to model for the potential additional cost of certain adverse effects (e.g. renal impairment) in the candesartan group;35 (ii) we added costs related to adverse effects not leading to hospital admission e.g. hypotension, renal impairment (those leading to admission are accounted for in base-case scenario), namely one extra general practice visit for each patient requiring dose reduction or treatment discontinuation for an adverse effect or laboratory abnormality; (iii) we varied the length of hospital stay in the per diem analysis by ±20% and (iv) for the UK only, we discounted costs by 3.5%, as recommended by the National Institute for Clinical Excellence (NICE).

Statistical methods
Statistical analysis was performed using SAS software (version 8; SAS Institute Inc., Cary, NC, USA). The group mean approach was used to account for early dropouts and missing values, as recommended by Cook et al.36 Bootstrapping was used to test for significance.

Role of the funding source
The CHARM programme was funded by AstraZeneca PLC, who managed the data and whose representatives were involved in the data analysis. The economic analysis was planned by all authors and supervised by J.Mc.M. All data were independently checked by S.S. and J.Mc.M. All authors were involved in interpretation of the data and writing of the manuscript which was prepared independently of the sponsor.


   Results
Top
 Abstract
 Introduction
 Methods
 Results
Discussion
 Acknowledgements
 References
 
Candesartan was taken by 50.0% of randomized patients (n=3803) for 1050 (SD 318) days on average; the comparable figures for placebo (n=3796) were 50.0% and 1029 (SD 345) days. The mean daily doses of candesartan taken were 16.8 mg in CHARM-Alternative, 16.9 mg in CHARM-Added, and 19.0 mg in CHARM-Preserved (17.7 mg overall), based on all patients in the study over the whole study period.

Clinical effects of candesartan
Hospital admissions—all causes
The rates and number of hospital admissions in the overall CHARM programme and each component trial are shown in Table 2 and Figures 1 and 2. Overall, 63.8% of patients in the placebo (conventional treatment) group were admitted to hospital at least once for any reason. This when compared with 62.4% in the candesartan group [odds ratio (OR) 0.94, 95% CI 0.86–1.03, P=0.20]. The number of admissions per patient hospitalized was 2.96 in the placebo group when compared with 2.82 in the candesartan group (P=0.045). The average length of an individual admission was 8.9 days in the placebo group and 9.0 days in the candesartan group. The average number of days spent in hospital for admitted patients was 26.3 days in the placebo group and 25.2 days in the candesartan group. As a result, treatment with candesartan resulted in fewer hospital admissions (placebo 7182, candesartan 6691 or 1.060 compared with 0.853 admissions per year of follow-up, P=0.0001) and fewer days in hospital (placebo 63 681, candesartan 59 923; Table 2, P=ns). The number of days in hospital per patient-year of follow-up was 6.0 in the placebo group and 5.5 in the candesartan group (P=0.056).


View this table:
[in this window]
[in a new window]
  		Table 2 Hospital admissions (any cause) and patients hospitalized (for any cause) in CHARM (totals and means)

 

Figure 0161
View larger version (12K):
[in this window]
[in a new window]
  		Figure 1 Cumulative number of hospital admissions: CHARM-Alternative (top left), CHARM-Added (top right), CHARM-Preserved (bottom left), CHARM-Overall (bottom right).

 

Figure 0162
View larger version (23K):
[in this window]
[in a new window]
  		Figure 2 Rates of hospital admission for any cause, all cardiovascular reasons, HF only, and non-cardiovascular reasons. The numbers in parentheses refer to the P-values for the comparison of the candesartan and placebo groups.

 
Hospital admissions—specific causes
Table 3 and Figure 2 show the breakdown of patients admitted (and hospital admissions) by specific causes. candesartan reduced both the proportion of patients admitted (–20%) and the number of admissions (–28%) for worsening HF. There was also a trend towards a reduction in atrial tachyarrhythmias. Conversely, there was significant increase in hospitalizations for hypotension (though the absolute numbers for this adverse effect were small i.e. 49 patients in the placebo group and 98 in the candesartan group). These findings were consistent across the component trials except for ‘other’ cardiovascular admissions, which were fewer in the candesartan group in CHARM-Alternative and CHARM-Added, but numerically greater in the candesartan group in CHARM-Preserved. Examination of these miscellaneous admissions did not reveal an excess in any specific category of event.


View this table:
[in this window]
[in a new window]
  		Table 3 Hospital admissions by cause—number of admissions and number and proportion of patients admitted

 
Procedures and operations
Cardiovascular procedures are shown in Table 4. The number of procedures, other than cardiac catheterization, was small and did not differ between treatment groups.


View this table:
[in this window]
[in a new window]
  		Table 4 Number of cardiovascular procedures in CHARM

 
Cost of adjunctive candesartan treatment: per diem analysis
The costs of adding candesartan to conventional treatment (compared with placebo added to conventional treatment), based on the analysis using the per diem costs, are shown in Table 5.


View this table:
[in this window]
[in a new window]
  		Table 5 Daily per patient cost of treatment in CHARM—per diem analysis (standard errors within brackets)

 
For France and the UK, the cost of care in the candesartan group was slightly (1–2%) less, even taking into account the cost of candesartan. For Germany, the overall cost of care was slightly higher (4%) in the candesartan group.

The costs of adding candesartan to conventional treatment differed by component trial in the CHARM programme. For all countries, there was a net cost-saving in CHARM-Alternative (0.3–7% reduction in cost) and either no additional cost or a cost-saving in CHARM-Added (0–7% reduction). In CHARM-Preserved, there was a net increase in the daily cost of care for all countries (6–12% increase in cost).

Cost of adjunctive candesartan treatment: DRG analysis
Table 6 shows the results of this analysis in the same way as Table 5 did for the per diem analysis. Although the costs of procedures were higher in France and Germany (and bed-day costs lower) than in the UK, the results of the DRG analysis were very similar to those of the per diem analysis. There was a small increase (2–6%) in the net daily cost of care with candesartan for all three countries using the DRG approach. Again, for all the three countries, the estimated net costs with candesartan were generally less in the CHARM-Alternative and -Added trials than in CHARM-Preserved.


View this table:
[in this window]
[in a new window]
  		Table 6 Daily per patient cost of treatment in CHARM—DRG analysis (standard errors within brackets)

 
Cost–consequence analyses
Translating the clinical findings and the daily costs into annual estimates gives the cost–consequence analysis shown in Table 7. Adjunctive treatment with candesartan in CHARM-Alternative and CHARM-Added led to clinical benefits and to either cost-savings or a small additional annual cost, depending on trial and country. The less certain clinical benefit in CHARM-Preserved was obtained at a modest extra annual cost in all the three countries.


View this table:
[in this window]
[in a new window]
  		Table 7 Cost–consequence analysis of candesartan compared with placebo in the treatment of HF—clinical benefits and annual per patient saving/cost increase (95% CI)

 
Cost-effectiveness analyses
Table 8 summarizes the cost-effectiveness analyses of the two reduced LVEF CHARM trials. These results were calculated using only DRG costs as all scenarios with per diem costs were cost-saving in all countries. Using French DRG costs, candesartan was cost-saving in the reduced LVEF trials individually and combined. In Germany, the cost per LYG incremental cost-effectiveness ratio (ICER) was estimated to range from {euro}1427 (CHARM-Added) to {euro}3881(CHARM-Alternative). The results for the UK lay between those of France and Germany, i.e. candesartan was cost-saving in CHARM-Added and the ICER was greatest (but still relatively small) in CHARM-Alternative at {euro}2547 per LYG.


View this table:
[in this window]
[in a new window]
  		Table 8 Cost effectiveness of candesartan in the CHARM reduced LVEF trials (based on DRG costs)

 
Sensitivity analyses
Increasing the length of stay for non-cardiovascular admissions by 30% increased the cost per day in the candesartan group by 15–20%. As a result, candesartan was no longer cost-saving in any comparison.

Adding one GP visit for an adverse event or laboratory abnormality which led to a reduction in the dose of, or discontinuation of, candesartan resulted in an increase in daily costs of {euro}0.01–0.02.

Varying the length of hospital stay for all admissions by ±20% varied costs in both groups accordingly, but did not change the general conclusions.

Using a 3.5% discount rate did not change the UK results much. Total daily costs were reduced by about {euro}0.04–0.11 in each of the CHARM trials individually and overall, with a similar change in both treatment groups.


   Discussion
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
Acknowledgements
 References
 
It is widely accepted that HF accounts for ~2% of direct health care costs in more developed countries and up to 70% of these are expended on hospital admissions.9–14 Furthermore, the public health (and economic) burden of HF is widely perceived to be increasing because of the ageing of populations in developed countries (elderly individuals having a greater prevalence of cardiovascular disease) and improving survival from conditions increasing the risk of future HF (e.g. myocardial infarction).8,11 Consequently, it is hoped that new treatments for HF will not only improve symptoms and reduce mortality, but also decrease hospital admissions and, in doing so, cut costs. Given the relationship between the cost of hospitalization and the overall cost of HF to society, any treatment that substantially reduces hospital admission rates is likely to be cost-effective.10,14 That is precisely what we found in this pre-planned economic analysis of CHARM. A substantial reduction in the proportion of patients admitted with worsening HF (and an even more marked reduction in the number of such admissions), without any increase in length of stay, contributed to a reduction in the rate of admission (and hospital bed days) for any reason, though this overall reduction was more modest. This is because the full impact of the reduction in admissions for worsening HF was attenuated by increased survival in the candesartan-treated patients (who, therefore, spent more time at risk of hospital admission for other reasons). Nevertheless, the cost-savings accruing from even this modest reduction in the rate of hospital admission for any cause largely offset the cost of candesartan. Though the final result varied slightly between the countries studied and according to the method of analysis (per diem compared with DRG), candesartan was, essentially, cost-neutral in the overall-CHARM programme (though clinical effectiveness was not proven in one component trial, CHARM Preserved). There was, however, heterogeneity between the component trials in the programme. Although candesartan treatment was associated with either a small reduction or increase in the net overall cost of care in CHARM-Alternative and CHARM-Added, depending on the analysis, in CHARM-Preserved (in which candesartan treatment did not reduce the primary endpoint significantly) there was a consistent and modest increase in the net cost of care. There appear to be two reasons for this. Though the proportional reduction in the rate and number of admissions for worsening HF was similar in all three CHARM trials, the absolute number of admissions prevented was smaller, relative to the number of patients treated, in CHARM-Preserved (i.e. the rate of admission for worsening HF was lower in CHARM-Preserved, Figure 2). Consequently, the cost-offset was less in CHARM-Preserved than in the other two trials. A second explanation was the increased number of ‘other’ cardiovascular admissions in the candesartan group in CHARM-Preserved (a reduction, rather than excess, of these admissions was observed in the other CHARM trials). A similar increase was observed with cardiovascular procedures in CHARM-Preserved. No clear pattern could be discerned in either excess suggesting that both increases may have been a chance finding. That the overall net cost of treatment in CHARM-Added was comparable with that obtained in CHARM-Alternative is notable, given that candesartan was added to full conventional treatment, including an ACE-inhibitor, in the former trial. It is also notable that the essentially cost-neutral outcome of these analyses of CHARM was obtained despite adding the cost of extra clinic visits and biochemical tests to reflect the extra costs related to initiating, up-titrating the dose, and monitoring the effects of candesartan.

Our findings on the cost–consequences of using candesartan are broadly in keeping with economic analyses of other effective treatments for HF though difficult to compare directly.22,37–39 No other placebo-controlled study included such a broad spectrum of patients, had within-trial data for such a long period of follow-up (with the exception of the ACE-inhibitor enalapril in the treatment arm of the studies of LV dysfunction in the case of the latter36), or added the test drug to such extensive background treatment. Nevertheless, a consistent message from these prior economic analyses and ours is that reduction in hospital admission offsets the cost of treatment. Remarkably, the cost-offset has been sufficient with all treatments examined, to date, to be cost-saving or more or less cost-neutral. This is despite each new drug being used as an additional treatment and against a trend of falling lengths of hospital stay.

These economic results, coupled with the clinical findings of the CHARM programme, have clear implications for the management of patients with HF.17–21,40 Not only does candesartan improve all important clinical outcomes in HF but also offers these benefits at little or no additional cost to the health care system; indeed, its use in patients with HF and reduced LV systolic function may lead to an actual reduction in the direct costs of health care. This is an important finding for health-care providers and society more generally, because there is no trade-off between the interest of the individual patient (and it seems unlikely that many patients would prefer the outcomes expected without candesartan) and the greater population served by the health-care system.

As with any analysis of this type there were limitations. By using the full unit cost of candesartan, our analyses have reduced the cost-effectiveness of this treatment for countries such as France where the patient or a private insurer pays 35% of the cost of treatment. We did not take account of indirect costs, such as loss of productivity due to inability to work; this, however, was unlikely to be a major problem because 54% of patients were beyond retirement age at the time of randomization. Pension payments in those who survived were not taken account of and there were more survivors in the candesartan group. We did not consider the cost related to death out of hospital, though there were more of these in the placebo group, this was a conservative approach. We did not incorporate quality of life information. We had less detailed and complete information on non-cardiovascular procedures and drugs. However, the main driver of costs is hospital admission and we did have information on these and tried to account for lack of information on the former in our sensitivity analyses. As with all economic analyses based on clinical trials of limited duration, there is concern that costs may only be postponed and that there may be ‘catch-up’ over the whole life-time of a patient. We believe that this is unlikely given the relatively long-duration of follow-up of CHARM (37.7 months) compared with the average life-expectancy of patients with HF. Another question about economic analyses is how far they can be generalized, geographically. We chose to focus on France, Germany, and the UK because robust national costs are available from government sources in these countries. We found very similar net consequences in each country and this finding and prior analyses suggest that broadly similar conclusions can be anticipated in other countries, given that in all countries, the cost of HF is mainly driven by hospital admissions. Clearly, however, a greater unit cost of candesartan, shorter (or otherwise less expensive) hospital stays or, especially, both, would make the cost outcome less favourable. Finally, we carried out a cost–consequence analysis of CHARM-Preserved even though the pre-specified primary outcome was not reduced significantly.

In summary, when added to currently recommended treatment, candesartan improves functional limitation due to symptoms,40 reduces hospital admissions for worsening HF17–20 and increases survival in patients with HF and a low LVEF20,21 and does this at little or no extra direct cost to the health-care system. candesartan is, therefore, a clinically and economically attractive adjunctive treatment for those patients, representing a ‘win-win’ scenario for both the individual patient and health-care providers.


   Acknowledgements
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 Acknowledgements
References
 
We are particularly grateful to all the investigators in CHARM who collected the data for these analyses. This paper was guest edited by Professor Peter L. Thompson, University of Western Australia, Sir Charles Gairdner Hospital, Perth, Nedlands, Australia.

Conflict of interest: none declared.


   References
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 Acknowledgements
 References
 

  1. Jessup M and Brozena S. (2003) Heart failure. N Engl J Med 348:2007–2018.[Free Full Text]
  2. McMurray JJV and Pfeffer MA. (2004) The year in heart failure. J Am Coll Cardiol 44:2398–2405.[Free Full Text]
  3. Stewart S, MacIntyre K, MacLeod MM, et al. (2001) Trends in hospital activity, morbidity and case fatality related to atrial fibrillation in Scotland, 1986–1996. Eur Heart J 22:693–701.[Abstract/Free Full Text]
  4. Stewart S, MacIntyre K, Hole DJ, et al. (2001) More ‘malignant’ than cancer? 5-year survival following a first admission for heart failure. Eur J Heart Fail 3:315–322.[Abstract/Free Full Text]
  5. Schaufelberger M, Swedberg K, Koster M, et al. (2004) Decreasing 1 year mortality and hospitalization rates for heart failure in Sweden: data from the Swedish Hospital Discharge Registry 1988 to 2000. Eur Heart J 25:300–307.[Abstract/Free Full Text]
  6. Hobbs FD, Kenre JE, Roalfe AK, et al. (2002) Impact of heart failure and left ventricular systolic dysfunction on quality of life: a cross-sectional study comparing common chronic cardiac and medical disorders and a representative adult population. Eur Heart J 23:1867–1876.[Abstract/Free Full Text]
  7. Blackledge HM, Tomlinson J, Squire IB. (2003) Prognosis for patients newly admitted to hospital with heart failure: survival trends in 12,220 index admissions in Leicestershire 1993–2001. Heart 89:615–620.[Abstract/Free Full Text]
  8. Bleumink GS, Knetsch AM, Sturkenboom MC, et al. (2004) Quantifying the heart failure epidemic: prevalence, incidence rate, lifetime risk and prognosis of heart failure. The Rotterdam Study. Eur Heart J 25:1614–1619.[Abstract/Free Full Text]
  9. Stewart S, Jenkins A, Buchan S, et al. (2002) The current cost of heart failure to the National Health Service in the UK. Eur J Heart Fail 4:361–370.[Abstract/Free Full Text]
  10. Berry C, Murdoch DR, McMurray JJ. (2001) Economics of chronic heart failure. Eur J Heart Fail 3:203–291.[Abstract/Free Full Text]
  11. Stewart S, MacIntyre K, Capewell S, McMurray JJ. (2003) Heart failure and the aging population: an increasing burden in the 21st century? Heart 89:49–53.[Abstract/Free Full Text]
  12. Selke B, Brunot A, Lebrun T. (2003) Economic repercussions of cardiac insufficiency in France. Arch Mal Coeur Vaiss 96:191–196.[Web of Science][Medline]
  13. Rydén-Bergsten T and Andersson F. (1999) The health care costs of heart failure in Sweden. J Intern Med 246:275–284.[CrossRef][Web of Science][Medline]
  14. Lee WC, Chavez YE, Baker T, Luce BR. (2004) Economic burden of heart failure: A summary of recent literature. Heart Lung 33:362–371.[CrossRef][Web of Science][Medline]
  15. Swedberg K, Pfeffer M, Granger C, et al. (1999) Candesartan in heart failure: assessment of reduction in mortality and morbidity (CHARM)—rationale and design. J Card Fail 5:276–282.[CrossRef][Medline]
  16. McMurray J, Östergren J, Pfeffer M, et al. (2003) Clinical features and contemporary management of patients with low and preserved ejection fracture heart failure: baseline characteristics of patients in the Candesartan in Heart failure—Assessment of Reduction in Mortality and morbidity (CHARM) programme. Eur J Heart Fail 5:261–270.[Abstract/Free Full Text]
  17. Granger CB and McMurray Yusuf S, et al. (2003) Effects of candesartan in patients with chronic heart failure and reduced left ventricular systolic function and intolerant to ACE-inhibitors: the CHARM-Alternative Trial. Lancet 362:772–776.[CrossRef][Web of Science][Medline]
  18. McMurray JJV, Östergren J, Swedberg K, et al. (2003) Effects of candesartan in patients with chronic heart failure and reduced left ventricular systolic function treated with an ACE-inhibitor: the CHARM-Added trial. Lancet 362:767–771.[CrossRef][Web of Science][Medline]
  19. Yusuf S, Pfeffer MA, Swedberg K, et al. (2003) Effects of candesartan in patients with chronic heart failure and preserved left ventricular systolic function: the CHARM-Preserved Trial. Lancet 362:777–781.[CrossRef][Web of Science][Medline]
  20. Pfeffer MA, Swedberg K, Granger CB, et al. (2003) Effects of candesartan on mortality and morbidity in patients with chronic heart failure: the CHARM-Overall programme. Lancet 362:759–766.[CrossRef][Web of Science][Medline]
  21. Young JB, Dunlap ME, Pfeffer MA, et al. (2004) Candesartan in Heart failure Assessment of Reduction in Mortality and morbidity (CHARM) Investigators and Committees. Mortality and morbidity reduction with candesartan in patients with chronic heart failure and left ventricular systolic dysfunction: results of the CHARM low-left ventricular ejection fraction trials. Circulation 110:2618–2626.[Abstract/Free Full Text]
  22. CIBIS-II Investigators and Health Economics Group. (2001) Reduced costs with bisoprolol treatment for heart failure: an economic analysis of the second Cardiac Insufficiency Bisoprolol Study (CIBIS-II). Eur Heart J 22:1021–1031.[Abstract/Free Full Text]
  23. Programme médicalisé des systèmes d'information (PMSI). www.le-pmsi.fr and www.atih.sante.fr.
  24. Anonymous. (2004) G-DRG, German Diagnosis Related Groups. Siegburg Version 2.1.
  25. Deutsche Krankenhaus Gesellschaft. (2003) Ergebnisse der Krankenhaus-Barometer. Umfrage, Düsseldorf http://www.dkgev.de.
  26. Anonymous. (2003) Verband der privaten Krankenversicherung eV. Die private Krankenversicherung, Zahlenbericht 2002/2003. Köln.
  27. Netten A and Curtis L. (2003) Unit Costs of Health and Social Care 2002 (PSSRU, University of Kent, Canterbury).
  28. Department of Health. (2002) National Schedule of Reference Costs–NHS Trusts. Available fromwww.dh.gov.uk.
  29. Club inter-pharmaceutique (CIP). www.cip-club.org.
  30. ifap-Serviceinstitut für Ärzte und Apotheker GmbH. IfAp®Index Praxis. Bad Saarow. (2004) Version 01.07.2004.
  31. Anonymous. (2004) Monthly Index of Medical Specialities (MIMS) March 2004 (Haymarket Medical Publications Ltd, London).
  32. Prescription Pricing Authority. (2004) National Health Service England and Wales Drug Tariff March 2004 (The Stationery Office, London).
  33. Gold MR, Siegel JE, Russell LB, Weinstein M. (1996) Cost-effectiveness in Health and Medicine (Oxford University Press, New York).
  34. Mauskopf JA, Paul JE, Grant DM, Stergachis A. (1998) The role of cost-consequence analysis in healthcare decision-making. PharmacoEconomics 13:277–288.[CrossRef][Web of Science][Medline]
  35. White HD. (2003) candesartan and heart failure: the allure of CHARM. Lancet 362:754–755.[Web of Science][Medline]
  36. Cook J, Drummond M, Heyse J. (2004) Economic endpoints in clinical trials. Stat Meth Med Res 13:157–176.[Abstract/Free Full Text]
  37. Glick H, Cook J, Kinosian B, Pitt B, Bourassa MG, Pouleur H, Gerth W. (1995) Costs and effects of enalapril therapy in patients with symptomatic heart failure: an economic analysis of the Studies of Left Ventricular Dysfunction (SOLVD) Treatment Trial. J Card Fail 1:371–380.[Medline]
  38. Paul SD, Kuntz KM, Eagle KA, Weinstein MC. (1994) Costs and effectiveness of angiotensin converting enzyme inhibition in patients with congestive heart failure. Arch Intern Med 154:1143–1149.[Abstract/Free Full Text]
  39. Glick HA, Orzol SM, Tooley JF, Remme WJ, Sasayama S, Pitt B. (2002) Economic evaluation of the randomized aldactone evaluation study (RALES): treatment of patients with severe heart failure. Cardiovasc Drugs Ther 16:53–59.[CrossRef][Web of Science][Medline]
  40. O'Meara E, Solomon S, McMurray J, Pfeffer M, Yusuf S, Michelson E, Granger C, Olofsson B, Young JB, Swedberg K. (2004) Effect of candesartan on New York Heart Association functional class. Results of the candesartan in Heart failure: Assessment of Reduction in Mortality and morbidity (CHARM) programme. Eur Heart J 25:1920–1926.[Abstract/Free Full Text]

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 CirculationHome page
S. D. Solomon, J. Dobson, S. Pocock, H. Skali, J. J.V. McMurray, C. B. Granger, S. Yusuf, K. Swedberg, J. B. Young, E. L. Michelson, et al.
Influence of Nonfatal Hospitalization for Heart Failure on Subsequent Mortality in Patients With Chronic Heart Failure
Circulation, September 25, 2007; 116(13): 1482 - 1487.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow E-letters: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when E-letters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (6)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by McMurray, J. J.V.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by McMurray, J. J.V.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?