CEAAAA

Cnip

D,*leisc

M. Arfan Ikram, MD, PHD,* Gabriel P. Krestin, MD, PHD, Albert Hofman, MD, PHD,*

Inondic

FroNetNetSwiCalDeBosTheanamanservhav

Macce

Journal of the American College of Cardiology Vol. 58, No. 16, 2011 2Pub

DownloadeJacqueline C. M. Witteman, PHD,* M. G. Myriam Hunink, MD, PHD*

Rotterdam, the Netherlands; Basel, Switzerland; San Francisco, California; and Boston, Massachusetts

Objectives The aim of this study was to assess the (cost-) effectiveness of screening asymptomatic individuals at intermedi-ate risk of coronary heart disease (CHD) for coronary artery calcium with computed tomography (CT).

Background Coronary artery calcium on CT improves prediction of CHD.

Methods A Markov model was developed on the basis of the Rotterdam Study. Four strategies were evaluated: 1) currentpractice; 2) current prevention guidelines for cardiovascular disease; 3) CT screening for coronary calcium; and4) statin therapy for all individuals. Asymptomatic individuals at intermediate risk of CHD were simulated overtheir remaining lifetime. Quality-adjusted life years (QALYs), costs, and incremental cost-effectiveness ratioswere calculated.

Results In men, CT screening was more effective and more costly than the other 3 strategies (CT vs. current practice: 0.13QALY [95% confidence interval (CI): 0.01 to 0.26], $4,676 [95% CI: $3,126 to $6,339]; CT vs. statin therapy: 0.04QALY [95% CI: 0.02 to 0.13], $1,951 [95% CI: $1,170 to $2,754]; and CT vs. current guidelines: 0.02 QALY[95% CI: 0.04 to 0.09], $44 [95% CI: $441 to $486]). The incremental cost-effectiveness ratio of CT calciumscreening was $48,800/QALY gained. In women, CT screening was more effective and more costly than current prac-tice (0.13 QALY [95% CI: 0.02 to 0.28], $4,663 [95% CI: $3,120 to $6,277]) and statin therapy (0.03 QALY[95% CI: 0.03 to 0.12], $2,273 [95% CI: $1,475 to $3,109]). However, implementing current guidelines wasmore effective compared with CT screening (0.02 QALY [95% CI: 0.03 to 0.07]), only a little more expensive($297 [95% CI: $8 to $633]), and had a lower cost per additional QALY ($33,072/QALY vs. $35,869/QALY). Sen-sitivity analysis demonstrated robustness of results in women but considerable uncertainty in men.

Conclusions Screening for coronary artery calcium with CT in individuals at intermediate risk of CHD is probably cost-effectivein men but is unlikely to be cost-effective in women. (J Am Coll Cardiol 2011;58:1690701) 2011 by theAmerican College of Cardiology Foundation

asymptomatic individuals, primary prevention of cor-ary heart disease (CHD) is often based on the pre-ted 10-year risk of a CHD event. The Framingham

risk factors are widely adopted for this purpose (1,2).Guidelines on cardiovascular disease (CVD) preventionrecommend advice on a healthy lifestyle (e.g., smokingcessation, regular physical activity) for individuals with a

See page 1702

low CHD risk (10%, 10-year risk), supplemented bystatins, antihypertensive medication, and sometimes as-pirin for individuals at high CHD risk (20%, 10-yearrisk) (35). In individuals at intermediate risk (10%to 20%, 10-year risk), the decision to treat with drugsis generally only recommended when either serum cho-

m the *Department of Epidemiology, Erasmus Medical Center, Rotterdam, theherlands; Department of Radiology, Erasmus Medical Center, Rotterdam, theherlands; Institute for Clinical Epidemiology, University Hospital Basel, Basel,tzerland; Division of Cardiology, Department of Medicine, University ofifornia San Francisco School of Medicine, San Francisco, California; and thepartment of Health Policy and Management, Harvard School of Public Health,ton, Massachusetts. The study was funded by ZonMw, project number 62300047.funding source had no role in the design and conduct of the study; collection,

lysis, and interpretation of the data; preparation, review, or approval of theuscript; or decision to submit the manuscript for publication. Dr. Krestin hased as a consultant to GE Healthcare. All other authors have reported that theye no relationships relevant to the content of this paper to disclose.Comparative Effectiveness andof Computed Tomography ScreeCoronary Artery Calcium in Asym

Bob J. H. van Kempen, MSC,* Sandra Spronk, PHSuzette E. Elias-Smale, MD, MSC,* Kirsten E. F

011 by the American College of Cardiology Foundationlished by Elsevier Inc.lesthr

anuscript received March 29, 2011; revised manuscript received May 11, 2011,pted May, 14, 2011.

d From: http://content.onlinejacc.org/ on 07/28/2014Cardiac Imaging

ost-Effectivenessng fortomatic Individuals

Michael T. Koller, MD,hmann, MD, MPH,

ISSN 0735-1097/$36.00doi:10.1016/j.jacc.2011.05.056terol or blood pressure levels are above a definedeshold. In this group, performing a noninvasive test

mifroonco

scotheonmeare(acctre(losuficmestrare

CTcalTheffasyCH

F

S intermcS

1691JACC Vol. 58, No. 16, 2011 van Kempen et al.Octo

Downloadeght afford identification of those who could benefitm more aggressive treatment. Coronary artery calciumcomputed tomography (CT), quantified by the CT

ronary calcium score, is such a test (6,7).Recent studies have demonstrated that the CT calciumre is a strong predictor of CHD risk, independent ofFramingham risk factors (716). In fact, more than

e-half of the individuals originally classified at inter-diate risk, on the basis of the Framingham risk factors,

reclassified to the high-risk (20%) or low-risk10%) category when the calcium score is taken intoount (7,17). Accordingly, these individuals should beated more aggressively (high risk) or less aggressivelyw risk). The reclassification to another risk categoryggests that using CT might be beneficial but reclassi-ation by itself is insufficient evidence to justify imple-ntation (18,19). Studies, ideally clinical trials, demon-ating comparative effectiveness and cost-effectiveness

necessary.In the absence of clinical trials showing the benefit of

screening, an extensive evaluation of CT coronarycium scoring with observational data is warranted (20).e objective of this study was to assess the comparativeectiveness and cost-effectiveness of screening anmptomatic elderly population at intermediate risk forD for coronary calcium with CT.

igure 1 Strategy Diagram

chematic representation of the 4 alternative strategies modeled for an individual at

ber 11, 2011:1690701omputed tomography (CT) coronary calcium screening, current practice, current guidelines,BP systolic blood pressure.

d From: http://content.onlinejacc.org/ on 07/28/2014thods

e developed a Markov decisiondel with TreeAge for Healthre (TreeAge Pro 2009, Tree-e Software, Williamstown,

assachusetts) to analyze rele-t strategies in asymptomaticerly individuals at intermedi-risk for CHD. The model

ucture, model parameters, andta sources are briefly describedre. Details of the modelingumptions and parameter esti-tion are given in the Onlinependix.odel structure. The followingstrategies were considered

ig. 1):

1. Current practice. This strategy reflects the incidenceof CHD and non-CHD events of individuals atintermediate risk without any additional preventiveintervention, as observed in the Rotterdam Study, andis used as the reference strategy. Some individualswere treated at baseline with statins, antihypertensivemedication, or aspirin by their general practitioners

ediate risk for coronary heart disease (CHD):

Abbreviationsand Acronyms

CHD coronary heartdisease

CI confidence interval

CT computedtomography

CVD cardiovasculardisease

ICER incremental cost-effectiveness ratio

LDL low-densitylipoprotein

QALY quality-adjustedlife year

SBP systolic bloodpressure

Cost-Effectiveness of CT Calcium ScreeningMe

WmoCaAgMvaneldatestrdaheassmaApM4(Fand statin therapy. LDL low-density lipoprotein;

Data Included in the Markov Model on CHD Primary Prevention Strategies forAsymptomatic Elderly Men and Women Identified as Being at Intermediate Risk for a CHD EventTable 1 Data Included in the Markov Model on CHD Primary Prevention Strategies forAsymptomatic Elderly Men and Women Identified as Being at Intermediate Risk for a CHD Event

Parameters: Probabilities and Characteristics of Base-Case Value Men* Base-Case Value Women*

In

In

In

F

T

T

T

U

1692 van Kempen et al. JACC Vol. 58, No. 16, 2011Cost-Effectiveness of CT Calcium Screening October 11, 2011:1690701

DownloadeReclassification Groups (n 329) (n 247) Distribution Data Source Ref. #

dividuals reclassified to low-risk group

n (%) 101 (31% [26% to 35%]) 95 (38% [33% to 45%]) Beta Cohort study (17)

Observed 10-yr CHD risk 0.05 (0.02 to 0.12) 0.08 (0.04 to 0.16) Beta Cohort study (17)

Observed 10-yr stroke risk 0.05 (0.02 to 0.12) 0.12 (0.04 to 0.16) Beta Cohort study (17)

dividuals not reclassified

n (%) 148 (45% [40% to 50%]) 98 (40% [34% to 46%) Beta Cohort study (17)



dividuals reclassified to high-risk group

n (%) 80 (24% [20% to 29%]) 54 (22% [17% to 27%]) Beta Cohort study (17)



irst-year CHD-related mortality after a CHD event 0.18 (0.13 to 0.22) Beta Cohort study (22)

RR of recurrent CHD event 1.5 (1.13 to 1.9) Triangular Cohort study (26)

First-year stroke-related mortality after a stroke event 0.20 (0.15 to 0.26) Beta Cohort study (28)

RR of recurrent stroke event 2 (1.5 to 2.5) Triangular Cohort study (28)

Lifetime risk of developing cancer due to radiationassociated with CT coronary calcium scanning

0.00008 (0.00005 to 0.00012) 0.00020 (0.00014 to 0.00028) Uniform Simulation study (29)

1-year case fatality given cancer due to radiation risk 0.65 (0.61 to 0.73) 0.70 (0.63 to 0.78) Triangular Simulation study (27)

reatment effectiveness of statins

RR on incident CHD, statins vs. placebo 0.70 (0.61 to 0.81) Log normal Meta-analysis (24)

RR on incident stroke, statins vs. placebo 0.81 (0.71 to 0.93) Log normal Meta-analysis (24)

Cost of statins, yearly $570 Pharmacy reference (33)

RR of hepatitis, statins vs. placebo 3.51 6.5 Log normal Meta-analysis (42)

RR of myopathy, statins vs. placebo 1.53 1.53 Log normal Meta-analysis (42)

Expected cost of hepatitis episode $120 Cohort study (43)

Expected cost of myopathy episode $250 Cohort study (43)

Expected disutility of hepatitis episode 0.05 QALY Cohort study (43)

Expected disutility of myopathy episode 0.025 QALY Cohort study (43)

reatment effectiveness of antihypertensives

RR of incident CHD, antihypertensives vs. placebo 0.85 (0.81 to 0.89) Log normal Meta-analysis (30)

RR of incident stroke, antihypertensives vs. placebo 0.64 (0.56 to 0.73) Log normal Meta-analysis (30)

Cost of antihypertensives, yearly $670 Pharmacy reference (33)

reatment effectiveness of aspirin

RR of incident CHD, aspirin vs. placebo 0.82 (0.75 to 0.90) Log normal Meta-analysis (38)

RR of incident stroke, aspirin vs. placebo 0.95 (0.85 to 1.06) Log normal Meta-analysis (38)

Annual rate major extracranial bleeding 0.0010 (0.00096 to 0.0012) Log normal Meta-analysis (38)

RR of major extracranial bleeding 1.54 (1.30 to 1.82) Log normal Meta-analysis (38)

1-yr case fatality due to major extracranial bleeding 0.03 (0.02 to 0.04) Log normal Estimate (28)

Cost of aspirin, yearly $20 Pharmacy reference (33)

Synergy factor 1 (0.9 to 1.1) 1 (0.9 to 1.1) Uniform See text

tilities

Asymptomatic elderly 0.86 (0.85 to 0.86) 0.83 (0.83 to 0.84) Normal Survey (32)

Post-CHD event 0.76 (0.74 to 0.77) 0.67 (0.65 to 0.69) Normal Survey (32)

Post-major bleeding 0.72 (0.70 to 0.74) 0.67 (0.65 to 0.69) Normal Survey (32)

Post-stroke event 0.69 (0.67 to 0.71) 0.61 (0.59 to 0.63) Normal Survey (32)

Disutility due to CHD event 0.10 (0.13 to 0.08) 0.16 (0.20 to 0.12) Triangular Survey (32)

Disutility due to major bleeding 0.14 (0.10 to 0.17) 0.16 (0.20 to 0.12) Triangular Survey (32)

Disutility due to stroke event 0.19 (0.24 to 0.16) 0.22 (0.26 to 0.19) Triangular Survey (32)

Continued on next paged From: http://content.onlinejacc.org/ on 07/28/2014

quhebleCHCHstranindspe

CoT

C

07)

00)

50)

1)

Data symptorisk n useCalc a in parfor t prevaleRott nts, pro201

C


Downloadeconsidered to be reflected in the observed incidence ofCHD and stroke.

2. Current guidelines. This strategy, based on fullyimplementing the most recent guidelines on primaryprevention of CHD for individuals at intermediaterisk for CHD, implies giving lifestyle advice to all,statin therapy when baseline low-density lipoprotein(LDL) cholesterol exceeds 130 mg/dl (3.37 mmol/l)(4), and antihypertensive medication when baselinesystolic blood pressure exceeds 140 mm Hg (5). In asensitivity analysis, we lowered the LDL threshold to100 mg/dl (2.59 mmol/l).

3. CT calcium screening. In this strategy a CT scanwas performed to determine the coronary calciumscore, and the 10-year CHD risk was recalculated onthe basis of the Framingham risk factors and thecalcium score combined. Consequently, a number ofindividuals will be reclassified to the high-risk orlow-risk category. Individuals reclassified to the low-risk category received lifestyle advice and pharmaco-logical treatment if systolic blood pressure was above140 mm Hg (21) and/or plasma LDL levels were160 mg/dl (4.14 mmol/l) (4). Individuals who re-mained in the intermediate-risk category were treatedas recommended for individuals at intermediate risk,similar to strategy 2. Individuals reclassified to thehigh-risk group received lifestyle advice, statin ther-apy, and antihypertensive medication, irrespective oftheir baseline cholesterol and blood pressure levels. Inaddition, men received low-dose aspirin (80 to 100 mgdaily). For both the current guidelines and CT cal-cium screening strategy, we assumed that individualswho used any of the 3 drugs at baseline wouldcontinue to use them.

4. Statin therapy. For this strategy we assumed thateveryone not currently taking a statin would receive amoderate-dose statin and would be otherwise man-

ntinuedable 1 Continued

Parameters: Probabilities and Characteristics ofReclassification Groups

Base-Case Value Men*(n 329)

osts, $

Performing a CT coronary calcium score $105 ($85 to $125)

First-year costs of CHD event $21,233 ($17,196 to $25,2

First-year costs of stroke event $24,000 ($18,000 to $30,0

First-year costs of major extracranial bleeding $13,400 ($10,500 to $23,4

Annual follow-up post-CHD or stroke event $1,330 ($1,006 to $1,60

Adherence, reference case analysis 70% (40% to 100%)

included in the Markov model on coronary heart disease (CHD) primary prevention strategies for a10% to 20%). Risk group classification, associated risk prediction, and prevalence of medicatioification Study). Major bleeding is defined as hemorrhagic stroke, gastrointestinal bleeding. *Dathe triangular and uniform distributions. Risk group classification, associated risk prediction, anderdam Coronary Calcification Study). Individuals stayed in intermediate-risk group. Risk of eve0 U.S. dollars.T computed tomography; RR relative risk.

ber 11, 2011:1690701aged according to current practice. Although initi-ating statins in all individuals is not always considered ep

d From: http://content.onlinejacc.org/ on 07/28/2014feasible in all situations, it puts the CT calciumscreening strategy into a broader perspective, betweenthe least aggressive strategy (current practice) andfairly aggressive strategy (statin therapy), providing arange of possibilities for an individual at intermediaterisk of CHD (20). Conceptually, an even more ag-gressive strategy would be to treat everyone not onlywith statins but also with antihypertensive medicationand aspirin (in men). In a sensitivity analysis, wesubstituted the statin therapy strategy with this ag-gressive medical treatment strategy.

For each of the 4 strategies, the model kept track ofality of life, costs, and time spent in 1 of the followingalth states: 1) well; 2) after CHD event; 3) after majoreding; 4) after stroke event; 5) after stroke event andD event; 6) after stroke and major bleeding; 7) afterD event and major bleeding; 8) after CHD event and

oke event and major bleeding; 9) CHD or stroke death;d 10) non-CHD or nonstroke death. Each simulatedividual started out in the well state. Age- and sex-cific probabilities of non-CHD death, fatal and nonfatalocardial infarction, fatal and nonfatal major bleeding dueaspirin use, fatal and nonfatal stroke, and lethal cancere to radiation determined the transition to the other statesring each annual cycle. The time horizon was the remain-

lifetime of the simulated individuals.A CHD event was defined as any of the followingtcomes: nonfatal myocardial infarction, coronary arterypass graft, percutaneous coronary intervention, and CHDrtality. In sensitivity analysis, we repeated the analysis

th hard CHD events as outcome, consisting of nonfatalocardial infarction and CHD mortality. Stroke was

fined as ischemic, hemorrhagic, or undefined stroke onebral CT. Major bleeding due to aspirin therapy (definedextra cranial hemorrhage leading to substantial disability)s modeled as a secondary event.

Base-Case Value Women*(n 247) Distribution Data Source Ref. #

Triangular Official tariff See text

Gamma Cost study (45)

Gamma Cost study (56)

Gamma Cost study (31,34)

Gamma Cost study (46,56)

Uniform Expert opinion See text

matic elderly men and women identified as being at intermediate risk for a CHD event (10-yearat baseline are based on a prospective observational cohort study (the Rotterdam Coronaryentheses are 95% confidence intervals for the beta- and (log)normal distributions and rangesnce of medication use at baseline are based on a prospective observational cohort study (thebabilities, utilities, and cost estimates described for men are identical for women. Costs are

Cost-Effectiveness of CT Calcium Screeningmytoduduing

oubymowimydeceraswaAfter a CHD event, stroke event, or a major bleedingisode, individuals moved to the post-CHD-event,

potivAftheCHtredeCTDatraatu1 wRo2,0toqucarIningmoonmocalrisanprothaintorfactofou

strbato:pohahigindrissufstr

onnaexphelishEfanwamethethethe

pirble

merelintbassetadadan

higanLaCHpoInmuproanim(sefacleaof

wecocatwiforspepoanpremm

CaprianAdwaforingsecmeonstaPlediscosioco

1694 van Kempen et al. JACC Vol. 58, No. 16, 2011

Downloadest-stroke event, or post-major bleeding state, respec-ely, or the combined states if 2 or all 3 events occurred.ter a major bleeding episode, we assumed that aspirinrapy would be discontinued. In the case of a nonfatalD or stroke, individuals would be allocated medical

atment for secondary CVD prevention. Non-CHDaths included fatal cancer due to radiation associated with

scanning.ta sources. Effectiveness of treatment, cost data, andnsition probabilities were retrieved from published liter-re and primary data collection and summarized in Tableith their data sources (17,2234).tterdam Study and event rates. From 1997 onwards,28 participants in the Rotterdam Study underwent CTdetermine their coronary calcium score and were subse-ently followed for 9.2 years (median) (12,17,35). Primarye physicians were blinded for the findings on CT.

terobserver and intraobserver agreement on calcium scor-has been found to be excellent (36). Two regression

dels were developed to predict the 10-year risk of CHDthe basis of the Framingham risk factors (predictiondel 1) and on the basis of Framingham plus the coronarycium score (prediction model 2) (17). The Framinghamk factors included were: age, systolic blood pressure,tihypertensive medication, total and high-density lipo-tein cholesterol, diabetes, and current smoking (4). Moren 50% of the individuals classified as Framinghamermediate-risk patients were reclassified to either high-low-risk when CT coronary calcium was added as risktor and the C-statistic increased significantly from 0.720.76 (17). The net improvement in reclassification wasnd to be 0.14 (p 0.01).

After excluding individuals who had a history of CHD oroke before the CT coronary calcium scan, we used theseline Rotterdam Study data and the 2 prediction models

1) determine the baseline characteristics of the targetpulation; and 2) determine the proportion of Framing-m intermediate-risk individuals reclassified to low andh risk when the coronary calcium score was added. Of allividuals reclassified to the low-, intermediate-, or high-

k categories, we observed how many of them actuallyfered from a CHD or stroke event with survival analysisatified by sex (Table 1).Probabilities of having a non-CHD event were calculatedthe basis of age- and sex-specific mortality rates from

tional life tables of the general population (37). Lifeectancy was adjusted for quality-of-life with mean

alth-related quality-of-life weights on the basis of pub-ed data (Table 1) (32).fectiveness of treatment. The benefit of statin andtihypertensive treatment on CHD and stroke incidences obtained from meta-analyses and considered equal forn and women (24,30). On the basis of a recent update,re is evidence that elderly men benefit from aspirin

Cost-Effectiveness of CT Calcium Screeningrapy in primary prevention of CHD. For elderly womenre remains considerable controversy (3). Therefore, as-

Coco

d From: http://content.onlinejacc.org/ on 07/28/2014in treatment for primary prevention was, when applica-, only modeled in men (38).Treatment adherence is an important determinant of treat-nt benefit (39). Although we used intention-to-treat-basedative risk reductions on the basis of clinical trials, which takeo account adherence, the adherence rate in a population-ed intervention is less than that achieved in the controlledting of a trial. We assumed, on the basis of expert opinion,herence to treatment in our population to be 70% of theherence in the original trials for the reference case analysisd explored a range of 20% to 100% in a sensitivity analysis.For secondary prevention and primary prevention inh-risk individuals, statins, antihypertensive medication,

din menaspirin therapy, are combined. Wald andw (40) estimated the effect of combining medication for

D prevention, but their approach does not account forssible synergy or dyssynergy between the drugs (41).stead, we estimated the effect of combining drugs byltiplying the individual relative risks and multiplying theduct by a synergy factor, which we varied in sensitivity

alyses between 0.9 and 1.100.9 implying synergy, 1.0plying independent effects, and 1.10 implying dyssynergye Online Appendix for details). The range in the synergytor was chosen such that a combination of drugs was atst as effective as a single component of the combinationthe same drugs.Because we considered a population at intermediate risk,

accounted for the fraction of individuals that used (ambination of) statins, aspirin, or antihypertensive medi-ion at baseline. An individual using statins at baseline butth LDL cholesterol levels 160, 130, and 90 mg/dl

the low-, intermediate-, and high-risk categories, re-ctively, was assumed to switch to a higher dose or more

tent statin. The same was assumed for an individual usingtihypertensive medication at baseline and systolic bloodssure levels 140 mm Hg, 140 mm Hg, and 120

Hg for the 3 risk categories, respectively.We assumed that all individuals in the Rotterdam Coronarylcium Study received lifestyle advice consistent with currentmary care practice and therefore that the observed CHDd stroke event rates reflected this intervention.verse effects. Hemorrhagic stroke due to aspirin therapys accounted for in the odds ratios of net treatment benefitstroke from the meta-analysis. Extracranial major bleed-due to aspirin therapy was modeled explicitly as a

ondary event with probabilities on the basis of a recentta-analysis (23). Myopathy and hepatitis were modeledthe basis of a meta-analysis of the adverse effects of

tins (42). On the basis of a recent modeling study bytcher et al. (43), we calculated the expected costs andutilities of a myopathy and hepatitis episode, includingsts for associated complications such as hospital admis-n, workup, and mortality, weighted by the probability ofmplications.

October 11, 2011:1690701sts. Costs incorporated in the model included healthcarests and non-healthcare costs and were assessed from the

socweind

cosicaaftnomeinfcommeLDpoatostaassbin

con60for

3.75 (2

.8%)

.4%)

.5%)

9%)

1%)

70%)

.8%)

3%)

37%)

4%)

0%)

6%)

low-den

CoT

C

S

C

C

CostStraincr[CT]

Q


Downloadeietal perspective for the United States (Table 1). All costsre converted to the year 2010 with the consumer priceexes.

Healthcare costs included costs of diagnostic procedures;ts for personnel, materials, and equipment; costs for med-tion; costs for healthcare resource use in subsequent yearser an event; and costs for overhead. The costs for ancontrast cardiac CT were based on healthcare reimburse-nt rates in 2009. Medication costs were based on pricingormation from the 2009 Red Book (33), which were

parable with current prices for statins, antihypertensivedication, and aspirin. We assumed, on the basis of baselineL cholesterol, that 30% of our population would need a

tent and more expensive statin, such as rosuvastatin orrvastatin, and the remaining 70% could do with a generictin such as simvastatin. For antihypertensive medication weumed that everyone would need at least a thiazide, com-ed with an angiotensin II receptor blocker, angiotensin-

Baseline Characteristics of Study PopulationWith Initial Risk of CHD Between 10% and 20%Table 2 Baseline Characteristics of Study PopWith Initial Risk of CHD Between 10%

Variable Men (n

Age, yrs 70 (66

Body mass index, kg/m2 26.5 (24

Systolic blood pressure, mm Hg 144 (13

Diastolic blood pressure, mm Hg 78 (70

Total cholesterol, mg/dl [mmol/l] 222 (201240)

HDL cholesterol, mg/dl [mmol/l] 46 (3363) [1

LDL cholesterol, mg/dl [mmol/l] 146 (124165) [

Cholesterol-lowering medication 52 (15

Antihypertensive medication 87 (26

Antithrombotic agents 97 (29

Smokers

Never 29 (

Current 70 (2

Former 230 (

Diabetes mellitus 19 (5

Calcium score

0 11 (

1100 122 (

101400 79 (2

4011,000 64 (2

1,000 53 (1

Values are mean (interquartile range) or n (%).CHD coronary heart disease; HDL high-density lipoprotein; LDL

st, Clinical Effectiveness, and Cost-Effectiveness of Strategies forable 3 Cost, Clinical Effectiveness, and Cost-Effectiveness of St

Total Lifetime Costs*Quality-Adjusted

Life Expectancy* (yrs)

urrent practice 7,551 10.03

tatin therapy 10,276 10.12

urrent guidelines 12,184 10.14

T screening 12,228 10.16

, clinical effectiveness, and cost-effectiveness of strategies for asymptomatic men (mean age 70tegies ordered by increasing cost. *Future costs and life years were discounted at 3%/year. 2010

ber 11, 2011:1690701emental cost-effectiveness ratios compared with both current practice and the next-best strategy. Dominscreening) that yields higher effectiveness at a lower incremental cost-effectiveness ratio.ALY quality-adjusted life year.

d From: http://content.onlinejacc.org/ on 07/28/2014verting enzyme inhibitor, or calcium channel blocker in% of individuals (44). Medication costs were only accounted

in adherent individuals. In a sensitivity analysis, we usederic prices for statins and antihypertensive medication,

imated to be $160 yearly for generic statins and $300 fortihypertensive medication (33). For both strategies 2 and 3,accounted for the costs of obtaining the Framingham risk

tors by a general practitioner, including laboratory costs.ent-related costs included the costs of hospital stay, diag-stic workup, interventions, and rehabilitation during the firstr after an event and were assumed to reflect the average coster a nonfatal myocardial infarction, coronary artery bypassft, or percutaneous coronary intervention (31,34,45,46).n-healthcare costs included travel costs and patient timets.alysis. All authors agreed on the model structure and

ta input before performing the analyses to ensure anjective and unbiased analysis.

n20%

) Women (n 247)

74 (7178)

7) 28 (2531)

) 149 (135161)

76 (6982)

.26.2)] 240 (217232) [6.2 (5.66.8)]

1.4)] 50 (3954) [1.3 (1.11.4)]

.425.1)] 158 (135178) [4.1 (2.635.62)]

44 (17.8%)

117 (47.4%)

43 (17.4%)

124 (50%)

33 (13%)

90 (36%)

42 (17.0%)

16 (7%)

104 (42%)

65 (26%)

37 (15%)

25 (10%)

sity lipoprotein.

ptomatic Menies for Asymptomatic Men

mental Costs/QALY ($/QALY)*pared With Current Practice

Incremental Costs/QALY ($/QALY)*Compared With Next-Best Strategy

Reference Reference

30,278 30,278

42,118 Dominated

35,977 48,800

) who have an intermediate risk (10% to 20%, 10-year risk) of coronary heart disease (CHD).ollars. Because not all strategies are considered feasible in all situations, we have presented

Cost-Effectiveness of CT Calcium ScreeninggenestanwefacEvnoyeaaftgraNocosAndaob

Asymrateg

IncreCom

yearsU.S. dulatioand

329

73)

.828.

1155

85)

[5.7 (5

.2 (1.1ated by extended dominanceimplies that there is another strategy (computed tomography

blothesextenQucretiobecalwethebosecwetiocoan

costrstrstrICexlifefordogawi(49senanTa

senthesenrelcliind(puinchimass5%risdeass

theindysianwh

ou(52coguwicu

F

(AanC(Bye


DownloadeImportant baseline characteristics, such as lipid levels,od pressure, and medication use, were determined forcohort of individuals at intermediate risk, stratified by

. The number of individuals using a statin, antihyper-sive, or aspirin under each strategy was determined.ality-adjusted life years (QALYs), lifetime costs, in-mental cost-effectiveness ratios (ICER) (i.e., addi-nal costs divided by QALYs gained), and net healthnefit (QALYs minus [costs/willingness-to-pay]) wereculated for all strategies. Future costs and effectivenessre discounted, to take into account time preference, at

currently recommended U.S. discount rate of 3% forth costs and effectiveness (47,48). To take into accountond-order uncertainty, 100,000 independent samplesre drawn from each of the input parameter distribu-ns, generating outcome distributions for QALYs andsts for each strategy. Calculations were done for mend women separately.Strategies were first ordered according to increasingst. A strategy was considered dominated if anotherategy was both more effective and less costly. Aategy was considered extended dominated if anotherategy achieved more effectiveness at a lower ICER. TheERs were calculated, after eliminating dominated andtended dominated strategies, as the difference in meantime costs divided by the difference in mean QALYseach strategy compared with the next-best non-

minated strategy. We considered $50,000/QALYined as a commonly accepted threshold for the societalllingness-to-pay threshold for primary prevention51) and varied it between $15,000 and $100,000 insitivity analyses. For the reference case analysis, we

alyzed the model with input parameters as given inble 1.Extensive 1-way, 2-way, multi-way, and probabilisticsitivity analyses were performed with plausible ranges of

parameter values. In particular, we explored modelsitivity to drug costs, aspirin therapy in women, and the

ative risk of an event with aspirin therapy. Because somenicians would be reluctant to withhold therapy from anividual who starts out with a predicted risk of 11%tting him originally at intermediate risk) andafterlusion of coronary calciuma revised risk of 9% (putting

at low risk), we explored the effect of an alternativeumption in which treating individuals reclassified to the

to 10% risk category as individuals with intermediatek (10% to 20%) and checked whether the optimalcision would change. Reclassification probabilities for thisumption are presented in Online Table 4.Because the 2004 guidelines on the initiation of statinrapy include an optional cutoff value of 100 mg/dl forividuals at intermediate risk, we did an additional anal-

s with this cutoff value in the current guidelines strategy

Cost-Effectiveness of CT Calcium Screeningd the CT calcium screening strategy for the individualso remained in the intermediate-risk group.

sn

d From: http://content.onlinejacc.org/ on 07/28/2014Probabilistic sensitivity analysis was performed with thetcome distributions of 100,000 Monte Carlo simulations). We calculated the probability that CT screening was

st-effective, compared with current practice, currentidelines, and statin therapy strategies for varyingllingness-to-pay thresholds, which yielded acceptabilityrves.

igure 2 Cost-Effectiveness Plane

) Cost-effectiveness plane in men. Quality-adjusted life expectancy in yearsgainst total lifetime costs in U.S. dollars, for computed tomography (CT) coro-ary calcium screening, current practice, current guidelines, and statin therapy.urrent guidelines are (extended) dominated by CT coronary calcium screening.) Cost-effectiveness plane in women. Quality-adjusted life expectancy inars against total lifetime costs in U.S. dollars, for CT coronary calcium

October 11, 2011:1690701creening, current practice, current guidelines, and statin therapy. The CT coro-ary calcium screening is (extended) dominated by current guidelines.

Re

Reistwofacmeintsta23(O(frTasligcu(41praofdifcuscrscrcu(Ctotheinceff(Q$4Figextleaaginc$3$4scrpraincanCItopa

0ThCTbeliferateffSethrdrscralsthetheradwogubeincopgeeffwheffthetremecu

CT

Co AsymT rateg

IncreCom

C

S

C

C

Cost 74 yeaincr causecost ted by high

A


Downloadesults

ference case analysis. Review of the baseline character-ics of the cohort at intermediate risk demonstrated thatmen were older than men and had less favorable risktor levels, apart from smoking and calcium scores (Table 2). Inn, implementing current guidelines for all individuals atermediate risk led to a steep increase in the number oftin and antihypertensive users (from 12% to 75% and% to 64%, respectively), compared with current practicenline Table 5). In women, a similar pattern was observedom 15% to 87% and 52% to 84%, respectively) (Onlineble 6). Implementing the CT screening strategy results inhtly fewer statin users, compared with implementing

rrent guidelines in both men (69% vs. 75%) and women% vs. 87%). In men, statin users with either currentctice or CT screening had a higher expected 10-year riskCHD compared with nonusers (Online Table 7). Thisference disappeared between users and nonusers withrrent guidelines. In women, this was only the case for CTeening (Online Table 8). In men (Table 3), CT calciumeening was more effective and more costly compared withrrent practice (QALY gain: 0.13 [95% confidence intervalI): 0.01 to 0.26], cost-increase: $4,676 [95% CI: $3,126$6,339]), more effective and more costly than statinrapy (QALY gain: 0.04 [95% CI: 0.02 to 0.13], costrease: $1,951 [95% CI: $1,170 to $2,754]), and more

ective but slightly more costly than current guidelinesALY gain: 0.02 [95% CI: 0.04 to 0.09], cost increase:4 [95% CI: $441 to $486]). The cost-effective plane inure 2A shows that, in men, current guidelines areended dominated by CT screening, because the latterds to a higher expected quality-adjusted life expectancyainst a lower incremental cost-effectiveness ratio. Theremental cost-effectiveness ratio of statin therapy is0,278/QALY, and for CT calcium screening it is8,800/QALY gained (Table 3). In women (Table 4), CTeening was more effective and more costly than currentctice (QALY gain: 0.13 [95% CI: 0.02 to 0.28]; costrease $4,663 [95% CI: $3,120 to $6,277]), more effectived more costly than statin therapy (QALY gain: 0.03 [95%: 0.03 to 0.12], cost increase: $2,273 [95% CI: $1,475

st, Clinical Effectiveness, and Cost-Effectiveness of Strategies forable 4 Cost, Clinical Effectiveness, and Cost-Effectiveness of St

Total Lifetime Costs*Quality-Adjusted

Life Expectancy* (yrs)

urrent practice 8,553 9.26

tatin therapy 10,944 9.36

T screening 13,216 9.39

urrent guidelines 13,514 9.41

, clinical effectiveness, and cost-effectiveness of strategies for asymptomatic women (mean ageeasing cost. *Future costs and life years were discounted at 3%/year. 2010 U.S. dollars. Be-effectiveness ratios compared with both current practice and the next-best strategy. Dominaer effectiveness at a lower incremental cost-effectiveness ratio.bbreviations as in Table 3.

ber 11, 2011:1690701$3,109]), and less expensive but also less effective com-red with current guidelines (QALY loss: 0.02 [95% CI:

ing$5

d From: http://content.onlinejacc.org/ on 07/28/2014.03 to 0.07], cost savings: $297 [95% CI:$8 to $633]).e cost-effective plane in Figure 2B shows that, in women,

screening is extended dominated by current guidelines,cause the latter leads to a higher expected quality-adjusted

expectancy against a lower incremental cost-effectivenessio, and therefore, CT screening is not considered cost-ective in women.nsitivity analysis. In men, at a willingness-to-payeshold of $50,000/QALY, a slight dyssynergy between

ugs would change the optimal decision from CTeening to statin therapy (Table 5). This shift wouldo occur if treatment adherence dropped below 58%,

effect of aspirin therapy on CHD was less protective,cost of a CT scan rose above $200, or the risk of

iation-induced cancer increased more than 10-fold. Inmen, the optimal strategy changed from currentidelines to statin therapy in case of a slight dyssynergytween drugs. Strong protective effects of aspirin on theidence of CHD and/or stroke would change thetimal strategy to CT screening (Table 6). Usingneric drug prices made the CT screening more cost-ective in men, with an ICER of $24,675/QALY,ereas in women current guidelines became more cost-ective, with an ICER of $21,140/QALY. Substituting

statin therapy strategy with the aggressive medicalatment strategy did not change the optimal decision inn. In women the optimal decision switched from

rrent guidelines to aggressive medical treatment.Probabilistic sensitivity analysis demonstrated that, in men,

screening was cost-effective compared with current prac-e in the majority of simulations if the willingness-to-payeshold was above $50,000 (Fig. 3A). In women, even ather willingness-to-pay thresholds, CT calcium screeninguld be cost-effective in 20% of the simulations (Fig. 3B).

scussion

this study we evaluated the comparative effectiveness andst-effectiveness of CT coronary calcium screening within

framework of current CVD prevention guidelines. Inn, the incremental cost-effectiveness ratio for CT screen-

ptomatic Womenies for Asymptomatic Women

mental Costs/QALY ($/QALY)*pared With Current Practice

Incremental Costs/QALY ($/QALY)*Compared With Next-Best Strategy

Reference Reference

23,910 23,910

35,869 Dominated

33,073 51,400

rs) who have an intermediate risk (10% to 20%, 10-year risk) of CHD. Strategies ordered bynot all strategies are considered feasible in all situations, we have presented incrementalextended dominanceimplies that there is another strategy (current guidelines) that yields

Cost-Effectiveness of CT Calcium Screeningticthrhigwo

Di

Incothemewas just below the willingness-to-pay threshold of0,000/QALY, and small changes in assumptions changed

CTSosucadextrisillua mscrdrures

efftioCThiganinexpwolesgroin

meobtreforfewbeis friseffi

whmotrecotheturcotrewiremscrtre

OpT

S

T

R

R

C

A

O

D

L

T

Y

A N/A

N/A

OpT

S

T

R

R

C

A

O

D

L

T

Y

A

Abb


Downloadescreening from being cost-effective to not cost-effective.me of the assumptions could be considered plausible,h as a slight dyssynergy between drugs or a treatment

herence lower than 60%, whereas others were morereme (e.g., a more than 10-fold increase in radiationk). The uncertainty in optimal decision was furtherstrated by the acceptability curves, which showed that, ininor but substantial proportion of the simulations, CT

eening was not cost-effective. However, with genericg prices the ICER for CT screening dropped, and theult was more robust in sensitivity analysis.In women, CT screening was not found to be cost-ective, even after using a wide range of varying assump-ns, which included assumptions more favorable to the

calcium screening strategy by treating individuals in theher end of low risk (5% to 10% risk) more aggressively

d with more treat-prone LDL thresholds. The differencethe optimal decision between men and women can belained by the fact that, compared with men, moremen were reclassified to the low-risk group, leading tos aggressive treatment. Furthermore, within the low-riskup, the observed risk of CHD is higher in women thanmen, so the foregone benefit with less aggressive treat-

timal Strategy in 1-Way Sensitivity Analysis, With a Willingness-to-able 5 Optimal Strategy in 1-Way Sensitivity Analysis, With a W

Variable of Interest Range Assessed in Dete

ynergy factor 0

reatment adherence 0

R of aspirin treatment on CHD 0

R of aspirin treatment on stroke 0

ost of CT coronary calcium scan $2

nnual incidence of radiation-induced cancer 1/100,

utcome Ha

iscount rates 4% for costs

DL threshold for statin initiation 10

reatment of individuals reclassified to 5%10% Similar to in

early drug prices of statins and antihypertensives Generic

ggressive medical treatment instead of statin therapy

not applicable; RR relative risk; other abbreviations as in Tables 2 and 3.

timal Strategy in 1-Way Sensitivity Analysis, With a Willingness-to-able 6 Optimal Strategy in 1-Way Sensitivity Analysis, With a W

Variable of Interest Range Assessed in Deter

ynergy factor 0.9

reatment adherence 0.2

R of aspirin treatment on CHD 0.2

R of aspirin treatment on stroke 0.2

ost of CT coronary calcium scan $20

nnual incidence of radiation-induced cancer 1/100,0

utcome Har

iscount rates 4% for costs,

DL threshold for statin initiation 100

reatment of individuals reclassified to 5%10% Similar to int

early drug prices of statins and antihypertensives Generic

Cost-Effectiveness of CT Calcium Screeningggressive medical treatment instead of statin therapy N/A

reviations as in Tables 2, 3 and 5.

d From: http://content.onlinejacc.org/ on 07/28/2014nt is higher in women. The benefit of CT screening istained in the high-risk group, where individuals areated more aggressively compared with current guidelines

treatment of intermediate-risk individuals. Becauseer women were reclassified to high risk, the potential

nefit of CT screening is lower than in men. The balanceurther shifted because aspirin is prescribed in men at highk but not in women, due to controversy with regard to itscacy in primary prevention of CHD.The Adult Treatment Panel (ATP)-IV guidelines,ich will be published soon, are expected to recommendre aggressive statin treatment than the current statinatment guidelines. Our statin therapy strategy can bensidered quite aggressive and is likely to be similar to

anticipated ATP-IV recommendation, ensuring fu-e applicability of our results. Of note, when we

mpared CT screening with an even more aggressiveatment strategy, as we did in the sensitivity analysisth the medical treatment strategy, CT screening

ained cost-effective in men. This implies that CTeening does not simply put more individuals onatment but allocates treatment to individuals who arepected to benefit most.

f $50,000/QALY in Menness-to-Pay of $50,000/QALY in Men

tic Sensitivity Analysis Optimal Decision (Base Case: CT Screening)

Statin therapy if synergy factor 1.02

Statin therapy if adherence 0.58

Statin therapy if RR 0.84

CT screening

0 Statin therapy if cost of CT $200

/100 Statin therapy if radiation risk increases 10-fold

CT screening

for QALYs CT screening

l CT screening

diate risk CT screening

rices CT screening

CT screening

f $50,000/QALY in Womenness-to-Pay of $50,000/QALY in Women

c Sensitivity Analysis Optimal Decision (Base Case: Current Guidelines)

Statin therapy if synergy factor1.02

Current guidelines

CT screening if RR0.73

CT screening if RR0.62

Current guidelines

100 Current guidelines

Current guidelines

or QALYs Current guidelines

Current guidelines

iate risk Current guidelines

ices Current guidelines

October 11, 2011:1690701ex

Pay oilling

ministi

1.1

1.0

1.0

1.0

$600

001/

d CHD

1.5% f

mg/dl

ermed

drug prPay oilling

rminis

.91.1

.21.0

.21.0

.21.0

0$60

0001

rd CHD

, 1.5%

0 mg/d

terme

drug pMedical treatment

calfroorincthatherel

CTcostioGepaPoStintagcalweintananext10yeaficwa

difadwocoreflresthasimsimopthe

Co

Sccowono

ReHuwam.

REF

1.

2.

3.

4.

F

(Apecinlacin


DownloadeA number of cost-effectiveness reports on CT coronarycium scoring have previously been published but differedm our study in the strategies or target population consideredin that they dichotomized the calcium score rather thanluding the score in a risk prediction. These studies foundt cost-effectiveness of CT screening was highly sensitive to

igure 3 Acceptability Curve

) Acceptability curve in men. For varying willingness-to-pay thresholds, theroportions of simulations that demonstrated cost-effectiveness for each strat-gy, computed tomography (CT) coronary calcium screening, current practice,urrent guidelines, and statin therapy, are indicated. (B) Acceptability curvewomen. For varying willingness-to-pay thresholds, the proportions of simu-tions that demonstrated cost-effectiveness for each strategy, CT coronary cal-ium screening, current practice, current guidelines, and statin therapy, aredicated.

ber 11, 2011:1690701population screened and downstream costs (5355). Theatively high incremental cost-effectiveness ratio we found for

5.

d From: http://content.onlinejacc.org/ on 07/28/2014screening in men is comparable to results of othert-effectiveness studies on interventions for primary preven-n of CHD, such as the study by Pletcher et al. (43).neralizability of our findings is further supported by com-

rable reclassification data on coronary calcium found bylonsky et al. (7) in the multi-ethnic study of atherosclerosis.udy limitations. First, we focused on individuals atermediate risk, which implied individuals were on aver-e older than 69 years of age. Screening for coronarycium could potentially have value in other subgroups, but

explicitly chose to investigate CT screening in theermediate-risk group as advocated by recent guidelinesd current consensus. Second, the time horizon in ouralysis was the remaining lifetime. Therefore, we had torapolate the incidence of CHD beyond the available-year data, but few simulated individuals lived beyond 15rs. Finally, although we stratified by sex, further strati-

ation by different combinations of baseline risk factorss not possible due to a limited sample size.As with all models of screening and diagnostic tests, theferences between the 4 strategies in terms of quality-justed life expectancy were small. Even though, inmen, the results seem robustly unfavorable for the CT

ronary calcium screening strategy, the residual uncertaintyected in the acceptability curves indicates that furtherearch might be beneficial. In men, the results indicatedt CT screening was cost-effective in the majority ofulations. Nevertheless, in a substantial proportion ofulations in men, current guidelines or statin therapy was

timal compared with CT screening, indicating that fur-r research is necessary.

nclusions

reening for coronary artery calcium with CT is probablyst-effective in men at intermediate risk of CHD. Formen at intermediate risk for CHD, CT screening doest seem to be cost-effective.

print requests and correspondence: Dr. M. G. Myriamnink, Room Ee 21-40a, Erasmus Medical Center, Dr. Mole-terplein 40, 3015 GD, Rotterdam, the Netherlands. E-mail:[email protected].

ERENCES

Wilson PW, DAgostino RB, Levy D, Belanger AM, Silbershatz H,Kannel WB. Prediction of coronary heart disease using risk factorcategories. Circulation 1998;97:183747.DAgostino RB Sr., Vasan RS, Pencina MJ, et al. General cardiovas-cular risk profile for use in primary care: the Framingham Heart Study.Circulation 2008;117:74353.Force USPST. Aspirin for the prevention of cardiovascular disease:U.S. Preventive Services Task Force recommendation statement. AnnIntern Med 2009;150:396404.Grundy SM, Cleeman JI, Merz CN, et al. Implications of recentclinical trials for the National Cholesterol Education Program Adult

Cost-Effectiveness of CT Calcium ScreeningTreatment Panel III guidelines. Circulation 2004;110:22739.Pearson TA, Blair SN, Daniels SR, et al. AHA guidelines for primaryprevention of cardiovascular disease and stroke: 2002 update: Consen-

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.


Downloadesus Panel Guide to Comprehensive Risk Reduction for Adult PatientsWithout Coronary or Other Atherosclerotic Vascular Diseases.American Heart Association Science Advisory and CoordinatingCommittee. Circulation 2002;106:38891.Helfand M, Buckley DI, Freeman M, et al. Emerging risk factors forcoronary heart disease: a summary of systematic reviews conducted forthe U.S. Preventive Services Task Force. Ann Intern Med 2009;151:496507.Polonsky TS, McClelland RL, Jorgensen NW, et al. Coronary arterycalcium score and risk classification for coronary heart disease predic-tion. JAMA 2010;303:16106.Greenland P, LaBree L, Azen SP, Doherty TM, Detrano RC.Coronary artery calcium score combined with Framingham score forrisk prediction in asymptomatic individuals. JAMA 2004;291:2105.Kondos GT, Hoff JA, Sevrukov A, et al. Electron-beam tomographycoronary artery calcium and cardiac events: a 37-month follow-up of5635 initially asymptomatic low- to intermediate-risk adults. Circula-tion 2003;107:25716.Arad Y, Goodman KJ, Roth M, Newstein D, Guerci AD. Coronarycalcification, coronary disease risk factors, C-reactive protein, andatherosclerotic cardiovascular disease events: the St. Francis HeartStudy. J Am Coll Cardiol 2005;46:15865.Taylor AJ, Bindeman J, Feuerstein I, Cao F, Brazaitis M, OMalleyPG. Coronary calcium independently predicts incident prematurecoronary heart disease over measured cardiovascular risk factors: meanthree-year outcomes in the Prospective Army Coronary Calcium(PACC) project. J Am Coll Cardiol 2005;46:80714.Vliegenthart R, Oudkerk M, Hofman A, et al. Coronary calcificationimproves cardiovascular risk prediction in the elderly. Circulation2005;112:5727.LaMonte MJ, FitzGerald SJ, Church TS, et al. Coronary arterycalcium score and coronary heart disease events in a large cohort ofasymptomatic men and women. Am J Epidemiol 2005;162:4219.Church TS, Levine BD, McGuire DK, et al. Coronary artery calciumscore, risk factors, and incident coronary heart disease events. Athero-sclerosis 2007;190:22431.Budoff MJ, Shaw LJ, Liu ST, et al. Long-term prognosis associatedwith coronary calcification: observations from a registry of 25,253patients. J Am Coll Cardiol 2007;49:186070.Detrano R, Guerci AD, Carr JJ, et al. Coronary calcium as a predictorof coronary events in four racial or ethnic groups. N Engl J Med2008;358:133645.Elias-Smale S, Proena RV, Koller MT, et al. Coronary calcium scoreimproves classification of coronary heart disease risk in elderly: theRotterdam study. J Am Coll Cardiol 2010;56:140714.Cook NR, Ridker PM. Advances in measuring the effect of individualpredictors of cardiovascular risk: the role of reclassification measures.Ann Intern Med 2009;150:795802.Pencina MJ, DAgostino RB Sr., DAgostino RB Jr., Vasan RS.Evaluating the added predictive ability of a new marker: from areaunder the ROC curve to reclassification and beyond. Stat Med2008;27:15772, discussion 20712.Pletcher MJ, Pignone M. Evaluating the clinical utility of a biomarker:a review of methods for estimating health impact. Circulation 2011;123:111624.Chobanian AV, Bakris GL, Black HR, et al. The Seventh Report ofthe Joint National Committee on Prevention, Detection, Evaluation,and Treatment of High Blood Pressure: the JNC 7 report. JAMA2003;289:256072.Adabag AS, Therneau TM, Gersh BJ, Weston SA, Roger VL.Sudden death after myocardial infarction. JAMA 2008;300:20229.Berger JS, Roncaglioni MC, Avanzini F, Pangrazzi I, Tognoni G,Brown DL. Aspirin for the primary prevention of cardiovascular eventsin women and men: a sex-specific meta-analysis of randomizedcontrolled trials. JAMA 2006;295:30613.Brugts JJ, Yetgin T, Hoeks SE, et al. The benefits of statins in peoplewithout established cardiovascular disease but with cardiovascular riskfactors: meta-analysis of randomised controlled trials. BMJ 2009;338:b2376.de Boer MJ, van Hout BA, Liem AL, Suryapranata H, Hoorntje JC,

Cost-Effectiveness of CT Calcium ScreeningZijlstra F. A cost-effective analysis of primary coronary angioplastyversus thrombolysis for acute myocardial infarction. Am J Cardiol1995;76:8303.

48.

d From: http://content.onlinejacc.org/ on 07/28/2014Deckers JW, Goedhart DM, Boersma E, et al. Treatment benefit byperindopril in patients with stable coronary artery disease at differentlevels of risk. Eur Heart J 2006;27:796801.Einstein AJ, Henzlova MJ, Rajagopalan S. Estimating risk of cancerassociated with radiation exposure from 64-slice computed tomogra-phy coronary angiography. JAMA 2007;298:31723.Greving JP, Buskens E, Koffijberg H, Algra A. Cost-effectiveness ofaspirin treatment in the primary prevention of cardiovascular diseaseevents in subgroups based on age, gender, and varying cardiovascularrisk. Circulation 2008;117:287583.Kim KP, Einstein AJ, Berrington de Gonzalez A. Coronary arterycalcification screening: estimated radiation dose and cancer risk. ArchIntern Med 2009;169:118894.Law MR, Morris JK, Wald NJ. Use of blood pressure lowering drugsin the prevention of cardiovascular disease: meta-analysis of 147randomised trials in the context of expectations from prospectiveepidemiological studies. BMJ 2009;338:b1665.Lightowlers S, McGuire A. Cost-effectiveness of anticoagulation innonrheumatic atrial fibrillation in the primary prevention of ischemicstroke. Stroke 1998;29:182732.Nyman JA, Barleen NA, Dowd BE, Russell DW, Coons SJ, SullivanPW. Quality-of-life weights for the US population: self-reportedhealth status and priority health conditions, by demographic charac-teristics. Med Care 2007;45:61828.Rice T. Red Book 2009: Pharmacys Fundamental Reference (RedBook Drug Topics). 113 Edition. Montvale, New Jersey: ThomsonReuters, 2009.Saini SD, Schoenfeld P, Fendrick AM, Scheiman J. Cost-effectivenessof proton pump inhibitor cotherapy in patients taking long-term,low-dose aspirin for secondary cardiovascular prevention. Arch InternMed 2008;168:168490, discussion 1691.Hofman A, Breteler MM, van Duijn CM, et al. The RotterdamStudy: 2010 objectives and design update. Eur J Epidemiol 2009;24:55372.Elias-Smale SE, Odink AE, Wieberdink RG, et al. Carotid, aorticarch and coronary calcification are related to history of stroke: theRotterdam study. Atherosclerosis 2010;212:65660.World Health Organization. Life tables for WHO Member States.Available at: http://www.who.int/healthinfo/statistics/mortality_life_tables/en/. Accessed August 26, 2011.Baigent C, Blackwell L, Collins R, et al. Aspirin in the primary andsecondary prevention of vascular disease: collaborative meta-analysis ofindividual participant data from randomised trials. Lancet 2009;373:184960.Ho PM, Bryson CL, Rumsfeld JS. Medication adherence: its impor-tance in cardiovascular outcomes. Circulation 2009;119:302835.Wald NJ, Law MR. A strategy to reduce cardiovascular disease bymore than 80%. BMJ 2003;326:1419.Indian Polycap Study (TIPS), Yusuf S, Pais P, et al. Effects of apolypill (Polycap) on risk factors in middle-aged individuals withoutcardiovascular disease (TIPS): a phase II, double-blind, randomisedtrial. Lancet 2009;373:134151.Law M, Rudnicka AR. Statin safety: a systematic review. Am J Cardiol2006;97:52C60C.Pletcher MJ, Lazar L, Bibbins-Domingo K, et al. Comparing impactand cost-effectiveness of primary prevention strategies for lipid-lowering. Ann Intern Med 2009;150:24354.Ma J, Lee KV, Stafford RS. Changes in antihypertensive prescribingduring US outpatient visits for uncomplicated hypertension between1993 and 2004. Hypertension 2006;48:84652.Kauf TL, Velazquez EJ, Crosslin DR, et al. The cost of acutemyocardial infarction in the new millennium: evidence from a multi-national registry. Am Heart J 2006;151:20612.Tsevat J, Kuntz KM, Orav EJ, Weinstein MC, Sacks FM,Goldman L. Cost-effectiveness of pravastatin therapy for survivorsof myocardial infarction with average cholesterol levels. Am Heart J2001;141:72734.Gravelle H, Brouwer W, Niessen L, Postma M, Rutten F. Discount-ing in economic evaluations: stepping forward towards optimal deci-sion rules. Health Econ 2007;16:30717.

October 11, 2011:1690701Siegel JE, Weinstein MC, Russell LB, Gold MR. Recommendationsfor reporting cost-effectiveness analyses. Panel on Cost-Effectivenessin Health and Medicine. JAMA 1996;276:133941.

49. Corso PS, Hammitt JK, Graham JD, Dicker RC, Goldie SJ. Assessingpreferences for prevention versus treatment using willingness to pay.Med Decis Making 2002;22:S92101.

50. Hirth RA, Chernew ME, Miller E, Fendrick AM, Weissert WG.Willingness to pay for a quality-adjusted life year: in search of astandard. Med Decis Making 2000;20:33242.

51. Lee KK, Cipriano LE, Owens DK, Go AS, Hlatky MA. Cost-effectiveness of using high-sensitivity C-reactive protein to identifyintermediate- and low-cardiovascular-risk individuals for statin ther-apy. Circulation 2010;122:147887.

52. Myrium Hunink MG, Glasziou PP, Siegel JE, et al. Decision Makingin Health and Medicine: Integrating Evidence and Values. Cam-bridge, United Kingdom: Cambridge University Press, 2001.

53. OMalley PG, Greenberg BA, Taylor AJ. Cost-effectiveness of usingelectron beam computed tomography to identify patients at risk forclinical coronary artery disease. Am Heart J 2004;148:10613.

54. Rumberger JA. Cost effectiveness of coronary calcification scanningusing electron beam tomography in intermediate and high riskasymptomatic individuals. J Cardiovasc Risk 2000;7:1139.

55. Shaw LJ, Raggi P, Berman DS, Callister TQ. Cost effectiveness ofscreening for cardiovascular disease with measures of coronary calcium.Prog Cardiovasc Dis 2003;46:17184.

56. Earnshaw SR, Scheiman J, Fendrick AM, McDade C, Pignone M.Cost-utility of aspirin and proton pump inhibitors for primary pre-vention. Arch Intern Med 2011;171:21825.

Key Words: coronary heart disease y cost-effectiveness analysis y CTcoronary calcium y CT screening y primary prevention.

APPENDIX

For supplementary figures, tables, and text,please see the online version of this article.

1701JACC Vol. 58, No. 16, 2011 van Kempen et al.October 11, 2011:1690701 Cost-Effectiveness of CT Calcium Screening

Downloaded From: http://content.onlinejacc.org/ on 07/28/2014

Comparative Effectiveness and Cost-Effectiveness of Computed Tomography Screening for Coronary A ...MethodsModel structureData sourcesRotterdam Study and event ratesEffectiveness of treatmentAdverse effectsCostsAnalysis

ResultsReference case analysisSensitivity analysis

DiscussionStudy limitations

ConclusionsReferencesAppendix

CEAAAA

Documents

Transcript of CEAAAA