American Journal of Cardiology
Volume 99, Issue 7 , Pages 1013-1015, 1 April 2007

The Things to Come of SHAPE: Cost and Effectiveness of Cardiovascular Prevention

  • George A. Diamond, MD

      Affiliations

    • Division of Cardiology, Cedars-Sinai Medical Center, Los Angeles, California
    • David Geffen School of Medicine, University of California, Los Angeles, Los Angeles, California.
    • Corresponding Author InformationCorresponding author: Tel: 323-462-0600; fax: 323-467-7490.
    • Dr. Diamond is a member of the speakers— panels of Merck & Co., Inc, West Point, Pennsylvania; and Schering-Plough Corporation, Kenilworth, New Jersey.
    • ,
  • Sanjay Kaul, MD

      Affiliations

    • Division of Cardiology, Cedars-Sinai Medical Center, Los Angeles, California
    • David Geffen School of Medicine, University of California, Los Angeles, Los Angeles, California.

Received 16 October 2006; received in revised form 30 October 2006; accepted 30 October 2006.

Article Outline

It has been proposed recently that asymptomatic adults undergo routine screening for subclinical atherosclerosis using electron-beam computed tomography or carotid ultrasound as the basis for the pharmacologic prevention of cardiovascular events. The expected cost and effectiveness of this conditional test-treatment strategy are herein analyzed in comparison with the conventional conditional strategy, based on Framingham risk factor assessment, and with an unconditional treatment strategy that does not rely on testing. These analyses show that the unconditional treatment strategy, albeit more costly, is more effective and more cost-effective than either conditional testing strategy. In conclusion, greater attention should be paid to improving the population-wide delivery of and long-term adherence to proven preventive therapy than to the identification of “high-risk” treatment targets.

 

“You see, if it’s—a reality, it’s the most important thing in the world. And I haven’t an atom of proof. Not an atom. Do you–?”

H.G. Wells, The Shape of Things to Come

An ad hoc task force of physician-scientists recently proposed the routine screening of asymptomatic adults for subclinical atherosclerosis using electron-beam computed tomography or carotid ultrasound as the basis for the pharmacologic prevention of cardiovascular events.1 The proponents of this so-called Screening for Heart Attack Prevention and Education (SHAPE) strategy acknowledge the absence of direct empirical support for the proposal but say that it is cost-effective in comparison with current National Cholesterol Education Program (NCEP) guidelines.2 The published executive summary, however, contains an incomplete accounting of the proposal—s costs and benefits and does not specify the cost-effectiveness ratios themselves.1 Such details are important, because these ratios are viewed by many as the categorical imprimatur for socially responsible health policy.3

A recent analysis of an alternative screening strategy on the basis of myocardial perfusion scintigraphy used a simple “back of the envelope” calculation to show that an unconditional treatment strategy was more effective and less costly than the conditional imaging strategy.4 The purpose of this essay is to apply the same approach to an assessment of the SHAPE proposal in comparison with the NCEP guidelines2 and with unconditional treatment. As with the earlier presentation,4 this analysis addresses only the proposal at hand and is not intended as a comprehensive discourse on the general process of epidemiologic screening, the principles of which are discussed elsewhere.5

Table 1 lists the cost and effectiveness of the 3 cardiovascular prevention strategies at issue. In each case, the “treatment” is considered to be the administration of a fixed dose of a statin at a cost of $2/day ($720/year), based on 2006 average wholesale prices,6 and the “test” is either Framingham risk factor assessment at a 1-time cost of $100 or electron-beam computed tomography at a 1-time cost of $400. In the absence of testing or treatment, we expect about 500,000 new atherosclerotic events per year (100,000 fatal and 400,000 nonfatal) in the 50 million middle-aged subjects targeted by SHAPE, an event rate of 1%.7

Table 1. Performance of alternative strategies for cardiovascular prevention
MetricTreat AllNCEPSHAPE
Test targets050million50million
Treatment candidates50million15million12.5million
Expected events500,000250,000400,000
Expected deaths100,00050,00080,000
Prevented events150,00075,000120,000
LYEs900,000450,000720,000
Prevented deaths30,00015,00024,000
Life-years390,000195,000312,000
Testing cost0$5billion$20billion
Treatment cost$36billion$10.8billion$9billion
Gross cost$36billion$15.8billion$29billion
Treatment savings$15billion$7.5billion$12billion
Net cost$21billion$8.3billion$17billion

On the basis of previous prospective trials, we expect statin treatment to reduce events by about 30%, independent of a variety of baseline characteristics.8, 9 The unconditional treatment of all 50 million subjects will thereby prevent 150,000 events (30% of 500,000) at a gross cost of $36 billion annually (50 million treatment candidates times $720 per year). Because each of these events incurs approximately $100,000 in direct and indirect costs,7 the prevention of this number of events can be expected to save $15 billion annually (150,000 events times $100,000 per event). The net cost of the unconditional treatment strategy is therefore $21 billion ($36 billion minus $15 billion).

In contrast, the NCEP guidelines2 would have us screen all 50 million targets with Framingham risk factor assessment at $100 each and treat only those with 10-year risk >20%. We can expect that 30% of the target population (15 million subjects) will manifest this level of risk and that 50% of the events will occur in these subjects (250,000 events), an event rate of 1.7%.10 If we now treat only these relatively “high risk” candidates, we will prevent 75,000 events (30% of 250,000) at a gross cost of $15.8 billion: $5 billion for testing (50 million test targets times $100 per test) plus $10.8 billion for treatment (15 million treatment candidates times $720 per year). The prevention of this number of events will save $7.5 billion annually (75,000 events times $100,000 per event).7 The net cost of this strategy is therefore only $8.3 billion, $12.7 billion less than the unconditional strategy, but it prevents half as many events.

Assuming that each fatal event is equivalent to a loss of 13 life-years7, 11 and that each nonfatal event is equivalent to a loss of 4 life-years,7, 11, 12, 13 each event represents a weighted loss of 6 life-year equivalents (LYEs). The marginal cost-effectiveness of unconditional treatment, therefore, is approximately $28,000/LYE (the $12.7 billion difference divided by the 75,000 event difference divided by 6 LYEs per event). The unconditional treatment strategy is therefore twice as effective as the NCEP—s conditional testing strategy and, despite its greater cost, satisfies the putative $50,000/LYE threshold criterion for cost-effectiveness.3, 14

The SHAPE strategy, in contrast, would have us screen all 50 million targets with electron-beam computed tomography and treat only those with calcium scores greater than some threshold value.1 For example, we can expect that 25% of the target population (12.5 million subjects) will manifest calcium scores >100 and that 80% of the events will occur (400,000 events) in these subjects, an event rate of 3.2%.15 If we now treat only this new “high-risk” population, we will prevent 120,000 events (30% of 400,000) at a gross cost of $29 billion: $20 billion for testing (50 million test targets times $400 per test) plus $9 billion for treatment (12.5 million treatment candidates times $720 per year). The prevention of this number of events will save $12 billion annually (120,000 events times $100,000 per event).7 The net cost of the SHAPE strategy is therefore only $17 billion, $4 billion less than the unconditional strategy, but it prevents 30,000 fewer events. The marginal cost-effectiveness of unconditional treatment compared with SHAPE, therefore, is $22,000/LYE (the $4 billion difference divided by the 30,000 event difference divided by 6 LYEs per event). Thus, although the unconditional treatment strategy costs 23% more than the SHAPE strategy, it is also 25% more effective. As a result, unconditional treatment is somewhat more cost-effective in comparison with SHAPE than it is in comparison with the NCEP guidelines.2

SHAPE—s cost-effectiveness analysis did not consider these comparisons. Instead, it compared the NCEP strategy with the proposed SHAPE strategy and reported only that SHAPE was cost-effective relative to this comparison. In contrast, Table 2 compares all 3 strategies using the 4 measures of cost-effectiveness recommended by the Institute of Medicine.3 According to the last of these analyses, SHAPE prevents 45,000 more events than the NCEP strategy at $8.7 billion greater net cost, a marginal cost-effectiveness ratio of $32,000/LYE (the $8.7 billion net cost difference divided by the 45,000 event difference divided by 6 LYEs per event). Although this supports SHAPE—s conclusion regarding cost-effectiveness,1 the additional analyses show that conclusion to be rather restrictive in focus. Yes, the SHAPE strategy is effective and cost-effective compared with the NCEP strategy, but unconditional treatment is even more effective and more cost-effective than the SHAPE strategy.

Table 2. Cost-effectiveness of alternative strategies for cardiovascular prevention (in $1,000)
MeasureSHAPE Versus NCEPTreat All Versus NCEPTreat All Versus SHAPE
Gross cost per life saved$1,467$1,346$1,167
Gross cost per life-year saved$113$104$90
Gross cost per LYE saved$49$45$39
Net cost per LYE saved$32$28$22

Of course, different parameters would materially change this conclusion. If the cost of treatment were twice as great and the cost of testing only half as much, for example, the cost-effectiveness ratio for unconditional treatment compared with SHAPE would increase to >$200,000/LYE, but even then, the unconditional strategy would continue to prevent 30,000 more events. Alternatively, if the test were interpreted using a more discriminatory criterion—a calcium score >400, which is capable of identifying 10% of the population experiencing 50% of the events, for example15—SHAPE—s effectiveness would decrease by 38% (from 120,000 to 75,000 prevented events), while its net cost would be reduced by only 4% (from $17 billion to $16.3 billion). At more liberal thresholds (e.g., a calcium score >0), unconditional treatment becomes the dominant strategy, costing less and preventing more events than SHAPE.

Nothing in this analysis should be construed as a criticism of the good intentions underlying SHAPE—s approach to the prevention of atherosclerotic events. Nor can cost, effectiveness, or cost-effectiveness ratios alone be used to decide the superiority of any strategy over another. These issues are economic and political, not scientific. Nevertheless, the evidentiary support for the orthodox approach based on risk factor assessment and therapeutic titration is clearly deficient,16 and innovative practical alternatives are sorely needed. In contrast, the analyses here identify the realistic bounds on our expectations imposed by the underlying mathematics of the screening process and the importance of emphasizing overall costs and benefits rather than cost-effectiveness ratios per se in the evaluation of competing strategies.4 From this perspective, it might be better to view cardiovascular prevention, like immunization, as a logistical problem of delivery rather than a strategic problem of triage. If so, we should focus more on treating a broad spectrum of the general population rather than identifying a small number of “high-risk” targets through an expensive and inefficient process of risk stratification. The effectiveness of treatment might be further improved by the addition of other agents, such as aspirin and angiotensin-converting enzyme inhibitors, and extension to those aged >75 years.

Regardless of these details, however, long-term adherence to preventive therapy continues to be a vexing problem. A number of studies consistently document the poor adherence to statin therapy required of all these proposed preventive strategies. In a random-effects meta-analysis of 5 recent studies comprising 52,319 patients,17, 18, 19, 20, 21 adherence to statin therapy over 2 years averaged only 62% (95% confidence interval 56% to 68%). Preliminary reports suggest that knowledge of coronary calcium scores might enhance adherence to statin therapy.22 Although this has not been confirmed in a prospective randomized trial,23 the magnitude of benefit could be sufficient to make SHAPE superior to unconditional treatment. If adherence were improved from 62% to 78%, for example, this would cancel out the 30,000 event difference between SHAPE and unconditional treatment. Moreover, this magnitude of difference in adherence could be confirmed over 2 years in a formal clinical trial comprising a sample of <400 subjects, compared with >400,000 subjects needed to test the expected 8% difference in event rate over the same period of time (1.4% for unconditional treatment vs 1.52% for SHAPE, assuming 100% adherence). If the SHAPE proposal serves to encourage such innovative and practical solutions to the problem of cardiovascular prevention, along with responsible ways to pay for them, that could be its most lasting legacy.

Back to Article Outline

References 

  1. Naghavi M, Falk E, Hecht HS, Jamieson MJ, Kaul S, Berman D, et al SHAPE Task Force (From vulnerable plaque to vulnerable patient—part III: executive summary of the Screening for Heart Attack Prevention and Education (SHAPE) Task Force report). Am J Cardiol. 2006;98:2H–15H
  2. National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). Third report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III) (Final report). Circulation. 2002;106:3143–3421
  3. In:  Miller W,  Robinson LA,  Lawrence RS editor. Committee to Evaluate Measures of Health Benefits for Environmental, Health, and Safety Regulation (Valuing Health for Regulatory Cost-Effectiveness Analysis). Washington, District of Columbia: National Academies Press; 2006;
  4. Diamond GA, Kaul S, Shah PK. Screen testing: Cardiovascular prevention in asymptomatic diabetics. J Am Coll Cardiol. In press.
  5. Sackett DL, Haynes RB, Tugwell P. In: Clinical Epidemiology: A Basic Science for Clinical Medicine. Boston, Massachusetts: Little, Brown; 1985;p. 144–155302–310
  6. Huron Valley Physicians Association. HVPA cost comparison: statins/statin combos. 2006;Available at: http://www.hvpa.com/hvpa_rx_drug_cost_tables_files/statin_cost_comparisons_0606upd.pdf. June. Accessed July 27, 2006.
  7. American Heart Association. Heart Disease and Stroke Statistics—2006 Update. Dallas, Texas: American Heart Association; 2006;
  8. Randomised trial of cholesterol lowering in 4444 patients with coronary heart disease: the Scandinavian Simvastatin Survival Study (4S). Lancet. 1994;344:1383–1389
  9. Heart Protection Study Collaborative Group. MRC/BHF Heart Protection Study of cholesterol lowering with simvastatin in 20,536 high-risk individuals: a randomized placebo-controlled trial. Lancet. 2002;360:7–22
  10. Greenland P, LaBree L, Azen SP, Doherty TM, Detrano RC. Coronary artery calcium score combined with Framingham score for risk prediction in asymptomatic individuals. JAMA. 2004;291:210–215
  11. In: Statistical Abstract of the United States (Table 116. Deaths—life years lost and mortality costs by age, sex, and cause: 2000 and 2002). Washington, District of Columbia: U.S. Department of Commerce; 2006;p. 92
  12. Sturm JW, Donnan GA, Dewey HM, Macdonell RA, Gilligan AK, Srikanth V, et al. Quality of life after stroke (The North East Melbourne Stroke Incidence Study (NEMESIS)). Stroke. 2004;35:2340–2345
  13. Tsevat J, Goldman L, Soukup JR, Lamas GA, Conners KF, Chapin CC, et al. Stability of time-tradeoff utilities in survivors of myocardial infarction. Med Decis Making. 1993;13:160–165
  14. Garber A. Cost-effectiveness and evidence evaluation as criteria for coverage policy. Health Aff (Millwood). 2004;(suppl):W4-284–W4-289
  15. Arad Y, Goodman KJ, Roth M, Newstein D, Guerci AD. Coronary calcification, coronary disease risk factors, C-reactive protein, and atherosclerotic cardiovascular disease events (The St. Francis Heart Study). J Am Coll Cardiol. 2005;46:158–165
  16. Hayward RA, Hofer TP, Vijan S. Lack of evidence for recommended low-density lipoprotein treatment targets: a solvable problem. Ann Intern Med. 2006;145:520–530
  17. Kopjar B, Sales AE, Pineros SL, Sun H, Li YF, Hedeen AN. Adherence with statin therapy in secondary prevention of coronary heart disease in veterans administration male population. Am J Cardiol. 2003;92:1106–1108
  18. Mantel-Teeuwisse AK, Goettsch WG, Klungel OH, de Boer A, Herings RMC. Long term persistence with statin treatment in daily medical practice. Heart. 2004;90:1065–1066
  19. Caspard H, Chan AK, Walker AM. Compliance with a statin treatment in a usual-care setting: retrospective database analysis over 3 years after treatment initiation in health maintenance organization enrollees with dyslipidemia. Clin Ther. 2005;27:1639–1646
  20. Campione JR, Sleath B, Biddle AK, Weinberger M. The influence of physicians— guideline compliance on patients— statin adherence: a retrospective cohort study. Am J Geriatr Pharmacother. 2005;3:229–239
  21. Perreault S, Blais L, Dragomir A, Bouchard MH, Lalonde L, Laurier C, et al. Persistence and determinants of statin therapy among middle-aged patients free of cardiovascular disease. Eur J Clin Pharmacol. 2005;61:667–674
  22. Kalia NK, Miller LG, Nasir K, Blumenthal RS, Agrawal N, Budoff JM. Visualization coronary calcium is associated with improvements in adherence to statin therapy. Atherosclerosis. 2006;185:394–399
  23. O—Malley PG, Feurstein IM, Taylor AJ. Impact of electron beam tomography, with or without case management, on motivation, behavioral change, and cardiovascular risk profile: a randomized controlled trial. JAMA. 2003;289:2215–2223

PII: S0002-9149(06)02510-0

doi:10.1016/j.amjcard.2006.10.070

American Journal of Cardiology
Volume 99, Issue 7 , Pages 1013-1015, 1 April 2007

Article Tools