(Circulation. 1996;93:1774-1776.)
© 1996 American Heart Association, Inc.
Articles |
Cost of Prevention
The Case of Lipid Lowering
From the Division of Cardiology and Preventive Cardiology and Therapeutics Program, Hamilton Civic Hospitals Research Centre, McMaster University, Hamilton, Ontario, Canada.
Correspondence to Dr Salim Yusuf, Hamilton General Hospital, 237 Barton St E, Hamilton, Ontario, Canada L8L 2X2. E-mail yusufs@fhs.mcmaster.csu.ca.
Key Words: Editorials • cost-benefit analysis • epidemiology • prevention • cholesterol
 |
Introduction |
|---|
 Top Introduction References |
|---|
A vast body of epidemiological studies has demonstrated a continuous relationship between serum cholesterol levels and CHD whereby the risk of CHD increases across the entire range of cholesterol values. A 10% reduction in cholesterol level over a period of several decades has been associated with a 30% reduction in CHD incidence, after correction for a number of confounders and adjustments for measurement errors.1 Clinical trials not only have addressed the question whether decreases in cholesterol levels will reduce CHD but have demonstrated how soon such benefits may be observed. Furthermore, comparison of the magnitude of benefit of relatively short-term treatment in the trials (eg, a few years) versus the theoretical maximum of long-term differences predicted from the epidemiological studies (eg, several decades) provides a framework to assess the potential long-term benefits of sustained cholesterol reduction, well beyond the duration of the trials.
The first generation of lipid-lowering trials (Helsinki [gemfibrozil],2 WHO [clofibrate],3 LRC-CPPT [cholestyramine]4 ) used drugs that lowered cholesterol only modestly (about 10%); some of these drugs produced side effects, so that compliance with treatment was suboptimal, and the studies were consequently inadequately powered to provide reliable estimation of the effects of lipid lowering on mortality. Although at first glance, the results of these trials appeared to be inconclusive or even contradictory, a comprehensive overview of these trials in an epidemiological context1 provided a coherent result: the gains in CHD reduction were related to the degree of cholesterol lowering and the duration of the intervention.
More recently, with the availability of 3-hydroxy-3 methyl glutaryl
coenzyme A reductase inhibitors, the degree of
cholesterol lowering is about 25%, a substantial effect
that is 2.5 times that seen in the first-generation trials with
cholestyramine or the fibrates. This large effect, combined with an
increased duration of treatment in trials of high-risk populations,
resulted in much larger reductions in CHD rates that were statistically
significant. These trials also were able to demonstrate clear
reductions in total mortality5 6 because they were largely
confined to high-risk individuals, eg, secondary prevention plus
high cholesterol levels (4S: >6.7 mmol/L) or primary
prevention with very high cholesterol levels (West of
Scotland Prevention Study: >7.0 mmol/L). More recently, other
secondary prevention trials have included patients with lower baseline
cholesterol levels CARE <6.2 mmol/L,7
Post-CABG Trial: LDL <3.6 mmol/L8 ). The CARE trial, which
enrolled 4159 post-MI patients with a mean baseline
cholesterol value of 5.9 mmol/L (LDL of 4.0 mmol/L) at
initial screening, achieved a 20% reduction in total
cholesterol with pravastatin and demonstrated a
24% reduction in CV death and MI after 5 years.9 The
Post-CABG surgery study,8 which included 1351 patients
between 1 and 12 years after their initial surgery (mean, 4.5 years),
randomized patients to two strategies: moderate lowering of LDL
cholesterol (to a mean of 3.5 mmol/L) versus aggressive LDL
lowering (to a mean of 2.4 mmol/L). There was a 29% reduction in
relative risk in the patients who demonstrated substantial progression
of saphenous vein graft disease (P<.0001), an 18%
reduction in major CV events (P<.09), and about a 25% risk
reduction in the composite of major CV events and repeat
revascularization procedures (P<.02).
Collectively, these trials support the conclusion that
cholesterol-lowering therapies aimed at a reduction in
total cholesterol of approximately 20% to 25% produce
important reductions in CV events in patients with preexisting vascular
disease across a broad range of cholesterol values within
the relatively short time frame of 5 years. Although these reductions
in CV events are substantial, the greatest impact of
cholesterol lowering occurs in individuals with the highest
baseline cholesterol levels that are also associated with
the highest baseline risk (Figure
).
|
In this issue of Circulation, a cost-minimization analysis of 4S (the study with the highest baseline cholesterol and risk) is presented by Pedersen et al.10 Cost-minimization analyses do not consider the potential benefits of therapy in comparison with the costs of therapy; rather, they consider the healthcare utilization costs and assume that the clinical effects are equivalent to the alternative treatment strategy. The deciding issue to accept or reject a proposed therapy is based on net cost. This analysis10 used data that were collected, prospectively provides a reliable estimate of the approximate costs of treatment over the period of time of the study, and concludes that the drug costs of treatment are largely offset by savings that result from fewer hospitalizations for myocardial infarction, unstable angina, or revascularization procedures.
A number of additional issues are worth exploring in an attempt to
understand the full implications of this analysis. Conversion
of the 4S cost-minimization analysis to a
cost-effectiveness analysis suggests that treatment with a
"statin" prevents a major CV event at a 5-year cost of $11 640 and
prevents a death at a 5-year cost of $23 280. This calculation is
comparable to the estimates by Goldman et al11 indicating
that in secondary prevention, when 20 mg/d of lovastatin is
used in individuals with serum cholesterol >6.4 mmol/L,
the cost per year of life saved is $25 000. Compared with other
cost-effectiveness analyses of a variety of commonly
accepted treatments, this estimate is very favorable.12 A
previous analysis13 of the LRC-CPPT reported a
cost effectiveness of 
$237 400 per year of life saved when
cholestyramine was used in individuals with total
cholesterol >6.8 mmol/L. Clearly, all
cost-effectiveness analyses, as we would expect,
demonstrate more favorable cost-effectiveness ratios for
individuals who are at higher absolute risk of CHD events and who have
higher pretreatment cholesterol levels. Therefore, cost
effectiveness in patients with vascular disease but lower baseline
cholesterol values would not be expected to be as
favorable. Nevertheless, lipid lowering by use of a statin is likely to
be cost effective compared with other treatments for CV disease even
when the background risk is half that seen in the 4S trial. Perhaps the
only primary prevention strategy utilizing
cholesterol-lowering agents that will prove to be
relatively cost effective is for patients with an increased baseline
risk (by virtue of other risk factors, such as diabetes or
hypertension) in addition to high pretreatment cholesterol
levels.
However, these analyses may not provide a complete picture if we assume that the reduction of cholesterol and its relationship to the reduction of CHD events remains constant over time. First, in the 4S study and the other three studies cited above, there was little or no effect during the first year or two of treatment, but at the end of 5 years of follow-up, impressive differences emerged, and the survival curves continued to diverge. This suggests that with more prolonged treatment, benefits of sustained cholesterol lowering are likely to increase in subsequent years until the effect size reaches the theoretical maximal benefit predicted from a similar degree of cholesterol difference on the basis of the epidemiological studies. For example, on the basis of the epidemiological data, a 20% reduction in cholesterol levels over several decades could be expected to lead to a 60% reduction in CHD events, a difference that is larger than the benefit seen at the end of 4S and other studies. Second, the relative impact of a specific difference in cholesterol is age dependent, as the epidemiological data suggest that the greatest benefits of cholesterol reduction occur in the youngest age group and that the increased risk of CHD in relation to high serum cholesterol decreases with increasing age. For example, among individuals aged 30 to 40 years, a 10% reduction in cholesterol results in a >50% reduction in CHD, whereas a similar cholesterol difference in a 60- or 70-year-old results in a 20% difference in CHD.1 Although the relative treatment benefits may appear to be larger among younger individuals, this is counterbalanced by their lower baseline risk of CHD events. Therefore, the overall absolute risk reduction in a given individual will depend on the age of initiation of treatment, the duration of intervention, and the inherent risk of a future vascular event. The latter is only partly determined by an individual's cholesterol level and is also dependent on the presence or absence of other risk factors (eg, blood pressure, diabetes, and tobacco use) and the presence or absence of vascular disease. All these factors should be taken into account in decisions regarding initiation of lipid-lowering therapy.
A separate but fundamental conceptual issue is the distinction between the costs of prevention of an acute infectious disease compared with the costs of prevention of chronic disease.14 In the first scenario, an individual in whom an infectious disease death is avoided in childhood or young adulthood can be expected to have a lifespan that is close to that of the average population. Such individuals rarely have any disability and would be expected to be reasonably productive. By contrast, with chronic diseases, the disease is generally not totally prevented but rather is postponed, and many individuals will probably still succumb to it, albeit a decade or two later. In some cases, although survival is prolonged, some individuals may be partly disabled, and their ability to be productive may be limited. Although this is clearly humanely worthwhile, in any cost-effectiveness analysis of chronic diseases, one should consider that costs are delayed rather than reduced; ultimately, the overall cost (spread over a longer time period) may be substantial and greater than any short-term analysis may demonstrate.
A third fundamental issue relates to the relatively modest population impact of any strategy that is applied to a highly select group of high-risk individuals. Because the relationship of cholesterol to CHD is graded across a broad range, strategies that shift the distribution of the entire curve by even a modest amount will have a substantial benefit. For example, a shift in the mean cholesterol level of the population by 10% would prevent 30% of all CHD events, whereas lifelong treatment of all individuals with a cholesterol level above the 90th percentile with a statin (achieving a 20% to 25% cholesterol reduction) would only prevent about 15% to 20% of all CHD events. Therefore, although population strategies to lower cholesterol may have only a relatively modest impact on cholesterol in a specific subgroup, the benefits spread over the entire population are substantial. These arguments suggest that both approaches, a population-based "low-tech" approach and a selective approach of treating high-risk individuals with cholesterol-lowering drugs, are integral and important components of a comprehensive approach to prevention of CHD.
|
Selected Abbreviations and Acronyms |
|---|
|
|
Footnotes |
|---|
The opinions expressed in this editorial are not necessarily those of the editors or of the American Heart Association.
|
References |
|---|
|
TopIntroductionReferences |
|---|
-
Law MR, Wald NJ, Thompson SG. By how much
and how quickly does reduction in serum cholesterol
concentration lower risk of ischemic heart disease?
BMJ. 1994;308:367-373.
[Abstract/Free Full Text]
-
Frick MH, Elo O, Haapa K, Heinonen OP, Heinsalmi P,
Helo P, Huttunen JK, Kaitaniemi P, Koskinen P, Manninen V, et al.
Helsinki Heart Study: primary-prevention trial with
gemfibrozil in middle-aged men with
dyslipidemia—safety of treatment, changes in risk
factors, and incidence of coronary heart disease.
N Engl J Med. 1987;317:1237-1245. [Abstract]
-
Committee of Principal Investigators. A
co-operative trial in the primary prevention of ischemic
heart disease using clofibrate: report from the Committee of Principal
Investigators. Br Heart J. 1978;40:1069-1118.
[Free Full Text]
-
Lipid Research Clinics Program. The Lipid
Research Clinics Coronary Primary Prevention Trial results, I:
reduction in incidence of coronary heart disease.
JAMA. 1984;251:351-364. [Abstract]
-
Scandinavian Simvastatin Survival Study
Group. Randomised trial of cholesterol lowering in
4444 patients with coronary heart disease: the Scandinavian
Simvastatin survival Study (4S). Lancet. 1994;344:1383-1389. [Medline]
[Order article via Infotrieve]
-
Shepard J, Cobbe SM, Ford I, Isles CG, Lorimer AR,
MacFarlane PW, McKillop JH, Packard CJ. Prevention of
coronary heart disease with pravastatin in men with
hypercholesterolemia. N
Engl J Med. 1995;333:1301-1307.
[Abstract/Free Full Text]
-
Pfeffer MA, Sacks FM, Moye LA, Brown L, Rouleau JL,
Hartley LH, Rouleau J, Grimm R, Sestier F, Wickemeyer W, Cole TG,
Braunwald E. Cholesterol and Recurrent Events: a
secondary prevention trial for normolipidemic patients.
Am J Cardiol. 1995;76:98C-106C. [Medline]
[Order article via Infotrieve]
-
Campeau L, Hunninghake DB, Domanski MJ. Post-CABG
Trial: design and results. Reported at the American College of
Cardiology 45th Annual Scientific Session, Orlando,
Fla: March 24-26, 1996.
-
Braunwald E, Pfeffer M, Sacks F. CARE Trial Results.
Reported at the American College of Cardiology
45th Annual Scientific Session, Orlando, Fla: March 24-26, 1996.
-
Pedersen TR, Kjekshus J, Berg K, Olsson AG, Wilhelmsen
L, Wedel H, Pyörälä K, Miettinen T, Haghfelt T,
Faergeman O, Thorgeirsson G, Jönsson B, Schwartz JS, for the
Scandinavian Simvastatin Survival Study Group.
Cholesterol lowering and the use of healthcare
resources: results of the Scandinavian Simvastatin Survival
Study. Circulation. 1996;93:1796-1802.
[Abstract/Free Full Text]
-
Goldman L, Weinstein MC, Goldman PA, Williams LW.
Cost-effectiveness of HMG-CoA reductase inhibition for
primary and secondary prevention of coronary heart
disease. JAMA. 1991;265:1145-1151. [Abstract]
-
Kupersmith J, Holmes-Rovner M, Hogan A, Rovner D,
Gardiner J. Cost-effectiveness analysis in heart
disease, I: general principles. Prog Cardiovasc Dis. 1994;37:161-184. [Medline]
[Order article via Infotrieve]
-
Weinstein MC, Stason WB. Cost-effectiveness
of interventions to prevent or treat coronary heart
disease. Annu Rev Public Health. 1985;6:41-63. [Medline]
[Order article via Infotrieve]
-
Rose GA. The Strategy of
Preventive Medicine. New York, NY: Oxford University Press;
1992.
This article has been cited by other articles:
 |
|
 |
|
  J. Malik, V. Melenovsky, D. Wichterle, T. Haas, J. Simek, R. Ceska, and J. Hradec Both fenofibrate and atorvastatin improve vascular reactivity in combined hyperlipidaemia (fenofibrate versus atorvastatin trial -- FAT) Cardiovasc Res, November 1, 2001; 52(2): 290 - 298. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Visioli, L. Borsani, and C. Galli Diet and prevention of coronary heart disease: the potential role of phytochemicals Cardiovasc Res, August 18, 2000; 47(3): 419 - 425. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. M. Grundy, J. I. Cleeman, B. M. Rifkind, L. H. Kuller, and for the Coordinating Committee of the National Cho Cholesterol Lowering in the Elderly Population Arch Intern Med, August 9, 1999; 159(15): 1670 - 1678. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. M. Gotto Jr and S. M. Grundy Lowering LDL Cholesterol : Questions From Recent Meta-Analyses and Subset Analyses of Clinical Trial DataIssues From the Interdisciplinary Council on Reducing the Risk for Coronary Heart Disease, Ninth Council Meeting Circulation, March 2, 1999; 99 (8): e1 - e7. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. A. Jacobson, J. R. Schein, A. Williamson, and C. M. Ballantyne Maximizing the Cost-effectiveness of Lipid-Lowering Therapy Arch Intern Med, October 12, 1998; 158(18): 1977 - 1989. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Robson Information needed to decide about cardiovascular treatment in primary care BMJ, January 25, 1997; 314(7076): 277 - 277. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||