Intended for healthcare professionals

Rapid response to:

Editorials

A cure for cardiovascular disease?

BMJ 2003; 326 doi: https://doi.org/10.1136/bmj.326.7404.1407 (Published 26 June 2003) Cite this as: BMJ 2003;326:1407

Rapid Response:

QUESTIONABLE COST-EFFECTIVENESS OF STATINS FOR PRIMARY PREVENTION OF CARDIOVASCULAR EVENTS

The Polypill has been proposed by Wald and Law [1] as a medication for
universal use by people aged over 55 years. In the editorial comments about the
Polypill, Rodgers [2] addresses the issue of the price that could be recognised
to this medication and discusses the cost-effectiveness implications of this
pharmacological intervention. Both the original paper [1] and the
above-mentioned editorial [2] implicitly or explicitly rely on the assumption
that the benefits of all the components of the Polypill,  including the statin, are conclusively
demonstrated in terms of both effectiveness and cost-effectiveness.

A large number of
studies indicate that statins are effective and cost-effective for secondary
prevention of cardiovascular events [3]. However, their role in primary
prevention is more controversial. In the present analysis, we focused our
attention on the pharmacoeconomic studies published from 1995 to June 2003 that
evaluated statins for primary prevention. Extraction of these studies from the
PubMed MEDLINE data bank [syntax of the query: "(cost[titl] OR
economic[titl]) AND (simvastatin OR pravastatin OR atorvastatin OR fluvastatin
OR cerivastatin OR lovastatin OR statin*)"; time limits: from January 1995
to June 2003; query launched on 11 July 2003] yielded a preliminary database of
articles that were candidate for our analysis. Direct inspection by AM, BS, ST,
and MV of the abstracts and, when necessary, of the full text of the papers of
this preliminary database allowed us to identify a total of 18 original
pharmacoeconomic studies evaluating statins for primary prevention; 8 were not
sponsored by any pharmaceutical company (table 1), whereas the other 10 were sponsored
(table 2). The design of the analysis was similar across these 18 original
studies (i.e. either comparison or statins vs. no therapy or head-to-head
evaluations between statins). On the other hand, the pharmacoeconomic results
showed that different studies favoured different statins.

In the independent
reports, the inter-study variations in the cost per life year gained were
extremely wide (table 1); this indicates a profound uncertainty in the
pharmacoeconomic convenience of using statins for primary prevention.

The sponsored studies (table 2) raised another and more important
question, i.e. whether the sponsor had any influence on the pharmacoeconomic
results. For this purpose, we reviewed all of the studies listed in Table 2 and
we assessed the relationship, if any, between the study sponsor and the type of
pharmacoeconomic result. In more detail, we analysed each study to determine
whether or not the pharmacoeconomic results were in favour of the statin
produced by the study sponsor.  The last
column of table 2 shows the results of our analysis. In all of these studies
(10 cases out of 10; 100%), the pharmacoeconomic results were in favour of the
statin produced by the study sponsor (p=0.00098 by signs’ test).

Thus, our overview of
these data provides a very negative picture on the scientific value of the
pharmacoeconomic research in the area of primary prevention with statins. One
negative finding is that most of these studies (N=10) were directly sponsored
by the pharmaceutical industry while the independent ones were fewer (N=8). In
addition, the research question of most of these studies did not address the
point of contrasting statins vs no statins for primary prevention (the most
relevant one from scientific and practical viewpoints), but considered various
head-to-head comparisons aimed at determining which statin is more
cost-effective than the others. In these head-to-head comparisons between
different statins, while the underlying question was the same across the
studies (“which statin has the best cost-effectiveness profile for primary
prevention?”), the results were surprisingly different. This casts doubts on
the robustness of the pharmacoeconomic methodology of these studies (inasmuch
as their results were conflicting from one another in the absence of specific
explanations). Last but not least, all of the sponsored pharmacoeconomic
studies found a better profile for the statin produced by the study sponsor
thus showing a clear bias towards favouring a pre-specified statin; this finding,
supported by a high statistical significance, is clearly the main result of our
literature analysis.

To explain our
findings, one hypothesis is that the high level of uncertainty in the
pharmacoeconomics of statins for primary prevention (table 1) has contributed
to creating a context where the sponsored studies could be guided towards the
"desired" result. Anyhow, one problem with pharmacoeconomic studies
in general is that the methods for constructing the statistical variability
around the primary result are still far from being standardised;  so, interpreting these data can be extremely
difficult under certain circumstamces.

In conclusion, after
reading the paper by Wald and Law [1] and the editorial by Rodgers[2], we
wonder whether these authors have over-estimated the clinical and economic
evidence about primary prevention with statins. Our view is that more rigorous
data are still needed in this field (particularly as regards the
cost-effectiveness indexes) before one can propose an indiscriminate use of
these agents in people aged more than 55 years.

 

REFERENCES

 

1.     Wald NJ, Law MR. A strategy to reduce cardiovascular disease by more
than 80%. BMJ. 2003;326(7404):1419.

2.     RodgersA. A
cure for cardiovascular disease?
BMJ 2003; 326: 1407-1408.

3.     Johannesson M, Jonsson B, Kjekshus J, Olsson AG, Pedersen TR, Wedel H.
Cost effectiveness of simvastatin treatment to lower cholesterol levels in
patients with
coronary heart disease.
Scandinavian Simvastatin
Survival Study Group.N Engl J Med. 1997;336(5):332-6.

 

Table 1. Cost-effectiveness of statins for primary prevention: list of
the studies not sponsored by pharmaceutical companies.



First author, year of publication, and
reference country (in parenthesis)


 


Type of economic analysis


Comparative design


Source of effectiveness
data


Results of the
pharmacoeconomic analysis†


Lim et al. (2001) (Australia)


CEA

(with simulation model)


Pravastatin vs.

no therapy


Epidemiological data


$110,000 (men) and $87,000   (women)
per life year gained.
[3%]


Perreault et al. (2000)
(Canada)


CEA


Simvastatin
vs. pravastatin vs.  atorvastatin vs.
fluvastatin vs cerivastatin vs. lovastatin


Meta-analysis of RCTs


Reduction in LDL-C less costly with simvastatin, pravastatin and
atorvastatin.

(men & women) [NR]


Pharoah and Hollingworth (1996)

(UK)


CEA

(with simulation model)


A statin  vs.

no therapy


1 RCT (WOSCOPS trial)


£136,000 per life year gained.

(men) [5%]


Pickin et al. (1999)

(UK)


 


CEA

(with simulation model)


A statin  vs.

no therapy


1 RCT (WOSCOPS trial)


From £8,200 to £12,500 per life year gained.

(men) [6%]


Prosser et al. (2000) (USA)


CEA

(with simulation model)


A statin  vs.

low-fat diet


Published data


From $54,000 to $420,000 (men) and from  $62,000 to $1,400,000 (women) per quality-adjusted life year
gained. 
[3%]


Rindone
and Arriola (1998)  (USA)


 


CMA


Fluvastatin  vs. simvastatin (design
based on the therapeutic switch from fluvastatin to simvastatin)


Observational study §§


Mean saving of $120 per patient per year using this switch.

(men & women)

[NR]


Spaans et al. (2003) (Canada)


CEA

(with simulation model)


A statin vs. no therapy


1 RCT plus a local cohort (with 1% of pre-existing coronary disease)


From $Can 7,700 to $Can 11,800 per life year gained (men). [3%]


Spearman  et al. (1997)

(USA)


 


CEA

(including also indirect
costs)


Fluvastatin or

lovastatin or

pravastatin or

simvastatin vs.

 no therapy


Observational study §§


Cost per 1% reduction of LDL-C lower for fluvastatin ($8.60)  than for the other statins (from $19.93 to
$23.59).

(men & women)

[0%]


 

§ Unless otherwise indicated, all of these analyses have considered
exclusively direct costs. The complete reference list is presented in the
footonote.

§§ = This study included both patients receiving primary prevention and
patients receiving secondary prevention.

  The figures enclosed by
brackets show the annual discount rate employed in the analysis.

ABBREVIATIONS: CEA = cost-effectiveness analysis ; CMA =
cost-minimization analysis; LDL-C = low-density lipoprotein cholesterol; Can$ =
Canadian dollar; RCT =  randomised  controlled trial;  NR = not reported;  £ =
sterling pounds;  [6%],[5%],[3%] = value
obtained with a discount rate of 6% or 5% or 3% for both costs and benefits,
respectively; [0%]= undiscounted value; [NR]= The study does not indicate whether
or not report if costs and benefits were discounted.

FOOTNOTE:

-     
Lim SS, Vos T, Peeters A, Liew D, McNeil
JJ.Cost-effectiveness of prescribing statins according to pharmaceutical
benefits scheme criteria. Med J Aust. 2001;175(9):459-64.

-     
Perreault S, Levinton C, Le Lorier J. Efficacy
and cost of HMG-CoA reductase inhibitors in the treatment of patients with
primary hyperlipidemia. Can J Clin Pharmacol. 2000;7(3):144-54.  

-     
Pharoah PD, Hollingworth W. Cost effectiveness
of lowering cholesterol concentration with statins in patients with and without
pre-existing coronary heart disease: life table method applied to health
uthority population. BMJ. 1996;312(7044):1443-8.

-     
Pickin DM, McCabe CJ, Ramsay LE, Payne N, Haq
IU, Yeo WW, Jackson PR. Cost effectiveness of HMG-CoA reductase inhibitor
(statin) treatment related to the risk of coronary heart disease and cost of
drug treatment. Heart. 1999;82(3):325-32.

-     
Prosser LA, Stinnett AA, Goldman PA, Williams
LW, Hunink MG, Goldman L. Cost-effectiveness of cholesterol-lowering therapies
according to selected patient characteristics.
Ann
Intern Med. 2000;132(10):769-79. 

-     
Rindone JP, Arriola G. Conversion from
fluvastatin to simvastatin therapy at a dose ratio of 8 to 1: effect on serum
lipid levels and cost. Clin Ther. 1998;20(2):340-6.

-     
Spaans JN, Coyle D, Fodor G, Nair R,
Vaillancourt R, Grover SA, Coupal L. Application of the 1998 Canadian
cholesterol guidelines to a military population: Health benefits and cost
effectiveness of improved cholesterol management. Can J Cardiol.
2003;19(7):790-6. 

-     
Spearman ME, Summers K, Moore V, Jacqmin R,
Smith G, Groshen S. Cost-effectiveness of initial therapy with
3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitors to treat  hypercholesterolemia in a primary care setting
of a managed-care organization. Clin Ther. 1997;19(3):582-602.

 

 

Table 2. Cost-effectiveness of statins for primary prevention: list of
the studies sponsored by pharmaceutical companies.§

First author, year of publication, and reference country (in
parenthesis)

Type of economic analysis

Comparative design

Source of effectiveness data

Results of the pharmacoeconomic analysis†

Sponsor

Pharmacoeconomic results favouring the statin manufactured by the
sponsor (Y/N)

Badia et al. (1999)

(Europe)

 

 

CMA

Simvastin vs.
atorvastatin

1 RCT

Cost per patient
lower for simvastatin (Euros 429) than for atorvastatin (Euros 538).

(men & women)

[NR]

Merck

Y

Caro et al. (1997)

(UK)

 

 

CEA

(with simulation model)

Pravastatin vs.

no therapy

1 RCT (WOSCOPS
trial)

£20,375 per life
year gained.

(men) [6%]

Bristol Myers
Squibb

 

Y*

Caro et al. (2000) (Belgium)

CEA (with simulation model)

Pravastatin vs. no
therapy

1 RCT (WOSCOPS
trial)

Euros 12,500 [0%
for benefits] or Euros 29,900 [5%] per life year gained (men).

 

Bristol Myers
Squibb

Y

Huse et al. (1998)

(USA)

 

CEA

(with simulation model)

Atorvastatin

Pravastatin or
simvastatin, or fluvastatin or

lovastatin vs.

no therapy

Physician Desk
Reference + Framingham coronary heart disease risk score

Cost per life year
gained lower for atorvastatin than for the other statins.

Values found for atorvastatin:

-men: $16,795  to 
$44,030 without risk factors or $4,294 to $13,064 with risk factors;

-women: less
than  $48000  with risk factors or more than $160,000 without risk factors.

[3%]

Pfizer

Y

Koren et al. (1998)

(USA)

 

CMA

Atorvastatin vs.
simvastatin  vs. lovastatin vs.
fluvastatin

1 RCT §§

Cost per patient
lower for atorvastatin $1064 than for the other statins from $1471 to $1972.
(men & women)

[NR]

Parke Davis-Pfizer

Y

Lacour et al. (1998)

(Canada)

 

 

 

 

CMA

Pravastatin 80mg vs.

simvastatin 80mg vs.

lovastatin 80mg vs.

bexafibrate 80mg vs.

lovastatin 80mg vs.

simvastatin 40mg vs.

simvastatin 30mg vs.

simvastatin 20mg vs.

phenofibrate 200mg
vs.

fluvastatin 60mg

124 studies  (randomised or non-randomised)

Among the agents
that reduce the ratio of total cholesterol/HDL cholesterol by >30%,
fluvastatin (60mg/day),  simvastatin
(20mg/day), and phenofibrate have the best pharmacoeconomic profile.

(men & women)

[NR]

Fournier-Pharma

Y

McPherson et al.(2001)
(Canada)

CMA

Cerivastatin vs. branded
pravastatin

1RCT

Cost savings in
favour of cerivastatin. (men & women) [NR]

Bayer

Y

Morris and Godber
(1999) (Canada)

 

CEA

Atorvastatin or
fluvastatin or

Lovastatin or pravastatin
or simvastatin vs.

no therapy

116 studies
(integrated with the Framingham coronary heart disease risk score)

Cost  per 1% reduction of LDL-C lower for
fluvastatin and atorvastatin.

Cost per life year
gained lower for these same 2 statins (from 
Can$36,000 to Can$45,000). 

(men & women)

[6% ]

Novartis

Y

Russell et al. (2001) (Canada)

CEA  (with simulation model)

Atorvastatin vs.
simvastatin  vs. fluvastatin vs.
lovastatin vs.

no therapy

Epidemiological
data §§

Atorvastatin cost
per life year gained vs fluvastatin 
from Can$ 12,333 to Can$ 60,509 (men) and from  Can$10,128 to Can$15,902 (women).

Parke Davis-Pfizer

Y

Smith and McBurney
(2003) (USA)

CMA

Atorvastatin vs. fluvastatin
vs. pravastatin vs. simvastatin 

1 RCT §§

Reduction in LDL-C
less costly with atorvastatin than with the other statins. (men & women)
[NR]

Pfizer

Y

 

§ Unless otherwise indicated, all of these analyses have considered exclusively
direct costs. The complete reference list is presented in the footonote.

§§ = This study included both patients receiving primary prevention and
patients receiving secondary prevention.

  The figures enclosed by
brackets show the annual discount rate employed in the analysis.

* The pharmacoeconomic result of this study has been considered
favourable to the statin according to current international standards for
interpreting the cost per life year gained.

ABBREVIATIONS:  Same as in Table
1.

FOOTNOTE:

-     
Badia X, Russo P, Attanasio E. A comparative
economic analysis of simvastatin versus atorvastatin: results of the Surrogate
Marker Cost-Efficacy (SMaC) study. Clin Ther. 1999;21(10):1788-96.

-     
Caro J, Klittich W, McGuire A, Ford I, Norrie
J, Pettitt D, McMurray J, Shepherd J. The West of Scotland coronary prevention
study: economic benefit analysis of primary prevention with pravastatin. BMJ.
1997;315(7122):1577-82.

-     
Caro JJ, Huybrechts KF, De Backer G, De Bacquer
D, Closon MC.Are the WOSCOPS clinical and economic findings generalizable to
other populations? A case study for Belgium. The WOSCOPS Economic Analysis
Group. West of Scotland Coronary Prevention Study. Acta Cardiol.
2000;55(4):239-46.

-     
Huse DM, Russell MW, Miller JD, Kraemer DF,
D'Agostino RB, Ellison RC, Hartz SC. Cost-effectiveness of statins. Am J
Cardiol. 1998;82(11):1357-63.

-     
Koren MJ, Smith DG, Hunninghake DB, Davidson
MH, McKenney JM, Weiss SR, Schrott HG, Henley RW Jr, Tresh P, McLain RW,
Bakker-Arkema RG, Black DM. The cost of reaching National Cholesterol Education
Program (NCEP) goals in hypercholesterolaemic patients. A comparison of
atorvastatin, simvastatin, lovastatin and fluvastatin. Pharmacoeconomics.
1998;14(1):59-70.

-     
Lacour A, Derderian F, LeLorier J. Comparison
of efficacy and cost among lipid-lowering agents in patients with primary
hypercholesterolemia. Can J Cardiol. 1998;14(3):355-61.

-     
McPherson R, Hanna K, Agro A, Braeken A.
Canadian Cerivastatin Study Group Cerivastatin versus branded pravastatin in
the treatment of primary hypercholesterolemia in primary care practice in
Canada: a one-year, open-label, randomized, comparative study of efficacy,
safety, and cost-effectiveness. Clin Ther. 2001;23(9):1492-507. 

-     
Morris S, Godber E. Choice of
cost-effectiveness measure in the economic evaluation of cholesterol-modifying
pharmacotherapy. An illustrative example focusing on the primary prevention of
coronary heart disease in Canada.
Pharmacoeconomics.
1999;16(2):193-205.

-     
Russell MW, Huse DM, Miller JD, Kraemer DF, Hartz
SC.
Cost effectiveness of HMG-CoA reductase inhibition in
Canada. Can J Clin Pharmacol. 2001;8(1):9-16. 

-     
Smith DG, McBurney CR.  An economic analysis of the Atorvastatin
Comparative Cholesterol Efficacy and Safety Study (ACCESS). Pharmacoeconomics.
2003;21 Suppl 1:13-23. 

 

Competing interests:  
None declared

Competing interests:  

18 July 2003
Andrea Messori
Coordinator
Benedetta Santarlasci, Sabrina Trippoli, and Monica Vaiani
Laboratorio di Farmacoeconomia, c/o Pharmaceutical Service, Careggi Hospital, 50132 Firenze, Italy