JACC Vol. 22, No .

6
Novembr15,9
3 ':Wre-702

1697

SMOKING CESSA FION

C0st-E&CdVei , , -3s of a Smoking Ces ation progra

Myocardial

After

HARLAN M. KRUMHOLZ, MD, BRIAN J . COHEN, MD,* JOEL TSEVAT, MD, MPH,t
RICHARD C . PASTERN AK, MD, FACCJ MILTON C . WEINSTEIN, PHD*
New /even, Connecticut and Boston,

Objectives . The purpose of this study was to evaluate the

cost-effectiveness of a smoking cessation program initiated after
acute myocardial infarction .
Background. The value of allocating health care resources to
smoking cessation programs after myocardial infarction has not
been compared with the value of other currently accepted interventions,
. A model was developed to examine the cost .
Methods
effectiveness of a recently reported smoking cessation prograin
after an acute myocardial infarction . The cost was estimated by
considering the resources necessary to implement the program,
and the effectiveness was expressed as discounted years of life
saved . Years of life saved were estimated by modeling life expectancy using a single declining exponential approximation of life
expectancy based on data from published reports .
Results . The cost-effectiveness of the nurse-managed smoking

cessation program was estimated to be $220/year of life saved . In

a one-way sensitivity analysis, the cost-effectiveness of the program remained <$20,000/year of life saved if the program decreased the smoking rate by only 3/1,(MO smokers (baseline
as umption 26/10 smokers), or if the program cost as much as
$g,0 /smoker (baseline assumption $100) . In a two-way sensitivity analysis, even if the cost of the program were as high as
$2,000/participant, the cost-effectiveness of the program would be
<310,/year
of life saved so long as floe program helped an
additional 12 smokers quit for every 100 carolled .
Conclusions. Over a wide range of estimates of costs and
effectiveness, a nurse-managed smoking cessation program after
acute myocardial infarction is an extremely cost-effective intervention . This program is more cost-effective than beta-adrenergic
antagonist therapy after myocardial infarction .

More than 25 years after the publication of the first U

.S .
Surgeon General's report (1), cigarette smoking remains an
important contributor to worldwide morbidity and mortality
from coronary heart disease . Cigarette smoking is estimated
to be responsible for >100,000 deaths from coronary heart
disease each year in the United States alone (2) . Smoking
after an acute myocardial infarction is particularly hazardous . Smokers who have had an acute myocardial infarction
and continue smoking have a much higher mortality rate
than do those who stop smoking (3)
.
Taylor et al . (4) reported a suc es ful smoking ces ation
program for patients hospitalized with an acute myocardial
infarction that included meetings with nurses and a regular,

brief telephone follow-up after hospital discharge . However,

despite the success of this program, organized smoking
cessation programs with telephone follow-up for survivors of
acute myocardial infarction are not widely disseminated and
are not currently reimbursed by most third-party payers in
the United States . One impediment to directing resources
toward smoking cessation programs may be the perception
that these programs are not worth their costs . The purpose
of this analysis was to determine the cost-effectiveness of a
reported smoking cessation program for smokers hospitalized with an acute myocardial infarction and to compare it
with that of other medical therapies .

(JAmCol ardiol193,2

:1697-02)

Methods
From the Section of Cardiovascular Medicine, Yale School of Medicine,
New Haven, Connecticut ; Cardiovascular Division and the Division of
Clinical Epidemiology, Departments of Medicine, tBeth Israel Hospital and
Brig'aam and Women's Hospital, *Harvard Medical School and the L partment of Health Policy and Management, Harvard School of Public Health,
Boston, Massachusetts . This study was supported in part by Training Grant
LM704fromtheNationalLibrayofMedicne,Bthesda,Mryland
GrantHL0734fromtheNatiunlHeart,Lung,adBlodInstiue,Nationl
Institutes of Health, Bethesda, and by the Institute of Respiratory Diseases
Foundation, Sarasota, Florida .
ManuscriptevdMach4,193
; revised manuscript rec ived May 24,
.
193,aceptdJun1,93
Address for correspondence : Dr. Harlan M . Krumholz, Section of CardiovascularMedicne,YaleSchol fMedicne,3CedarSte ,New
Haven,Co ctiu0651
.
913bytheAmricanColeg fCardiolgy

Downloaded From: https://onlinejacc.org/ on 08/17/2014

e developed a model to compare a nurse-managed

smoking cessation program with usual smoking cessation
counseling for survivors of acute myocardial infarction . The
model is represented graphically as a decision tree in Figure
1 . The benefit was expressed as the number of discounted
life-years saved as a result of the expected reduction in
mortality among patients who stopped smoking . All calculations were rounded to the nearest $10 .
Description of the smoking cessation program . As described by Taylor et al . (4), the program is in tiated in the
hospital when a smoker becomes clinically stable after an
0735-19)6

.00

KRUMHOLZ ET AL .
SMOKING CESSATION AFTER MYOCARDIAL INFARCTION

1698

Figure 1 . A decision tree of the choice to refer a smoker for a

smoking cessation program after an acute myocardial infarction .

acute myocardial infarction . A nurse trained in smoking

cessation techniques visits the patient and reviews the risks
of continued smoking and the benefits of smoking cessation .
The nurse gives the patient a manual that explains how to
identify high risk smoking situations and counsels him or her
about how to cope with the temptation to smoke . After
hospital discharge, nurses cal the patients we kly for 3
we ks and then monthly for 4 months to provide sup ort
.
Probabilities. Taylor et al .(4)comparedtheirpogram
with usual postmyocardial infarction care for smokers, consisting of a firm, unequivocal message from doctors and
nurses to the patients to stop smoking, and reported that 71
of every 100 smokers in the intervention group stopped
smokingcomparedwith45ofevry10 smokersintheusal
care group, Subjects in the study were followed up for I
year, and their smoking status was verified by measurement
of expired carbon monoxide and by serum thiocyanate
levels, On the basis of that study, we assumed that the new
program would be responsible for helping an ad itional 26
smokers stop smoking for every 100 smokers enrolled in the

Novembr15,93

JACC Vol . 22, No .6

.1697-02

program (I person stop ing for ev ry 3

.8 smokers enrolled in
the program).
Several studies (Table 1) have observed the smoking
status of large groups of smokers (> 100 subjects) after an
acute myocardial infarction and then followed them up for
>5years(-10)
. In thes studies the 5-year mortality rate
amongsmokers angedfrom2%to47%
. All of the large
studies have shown a substantial reduction in mortality in
patients who quit smoking compared with that of patients
who continued smoking . Aberg et al .(5),inthelargest udy,
folwedup983men<6yarsofge(man52
.6years)fo
up to 10 .5 years after acute myocardial infarction
. In that
study, smoking status was ascertained 3 months after infarction . At 5 years, the mortality rate was 16% in the quit ers
and 2 % in the smokers, a 27% relative dif erence
. This
difference in mortality rates was among the lowest in the
studies reviewed and is the estimate that we used in our
baseline analysis .
Life expectancy was modeled for smokers and nonsmokers using a single declining exponential curve (11) with a time
constant calculated from the 5-year mortality rates from the
. These curves fit the reported data
study of Aberg et al .(5)
well for at least the 1st 8 years of the study (Fig . 2) . The
modeled survival curve for the ex-smokers slightly underestimates the reported survival figures, whereas the modeled
survival curve for the smokers slightly overestimates survival . Discounted life expectancy was calculated for smokers and ex-smokers by taking the integral under each survival curve from year i to year i + I, multiplying that by
110 .051,andsumingfori=Ito50
. The resultant
calculated gain in life expectancy from smoking cessation
(life expectancy of ex-smokers minus life expectancy of
smokers) is estimated to be 1 .7years
.
Costs. The program was assumed to require approximately 3 h of nursing time p r
. The cost of nursing
r Atien(4)
time was estimated to be ap roximately $30/h, based on
estimated prevailing salaries in the Boston area . We assumed that each patient would be given a self-help manual
and other instructional material at a cost of approximately

Table 1 . Summary of Large Observational Studies of the Effect of Smoking Cessation on Mortality After Acute Myocardial Infarction
With at Least 5 Years of Fol ow-Up
5-YearMotaliyRate
Stud

Patients Studied (no.)

Sparow(8),197

Quitters

365(29ten,wom)

Smokers

Comment

12%

25%

Framingham Heart Study ; quitting defined as not smoking after first MI

on the biennial examination after MI

Abets(),1983

983(menoly)

16%

22%

Quit ing defined as not smoking 3 months after MI

Dal), (10) .1983

498(menoly)

20%

30

Quitting defined as not smoking 2 years after first MI or unstable

angina episode

Johans (6),19M

156(women ly)

15%

27%

Quit ing defined as not smoking 3 months after MI

Perkins(7),1985

19(0men,2wo)

21%

47%

Quit ing defined as not smoking at 3 months, 6 months and yearly for
upto5yearsof ol w-up

169'e

31%

Quitting not defined (smoking defined as regular smoking in the month

before M0

Hedback(9),187

305(28men,47wo)

*First author, reference number, year of publication . MI = myocardial infarction .

Downloaded From: https://onlinejacc.org/ on 08/17/2014

JACC Vol. 22, No .6

Novembr15,93
1:07-2

KRUMHOLZ ET AL .
SMOKING CESSATION AFTER MYOCARDIAL INFARCTION
to

1 .0
mmg od

0.9
IENALE

0.8

Figure 2 . Comparison of survival curves after

myocardial infarction (published values from
Aberg et al .15]andvaluescaluatedfrom
modeling a declining exponential approximation of life expectancy [DEALE]) for exsmokers (left) and smokers (right),

DEALE

0.7
0.6
-

as

05
0

$10/patient . Therefore, the estimated cost of the program for

each patient was $100 . Our analysis did not include the cost
of setting up the program and initially training the nurses .
To be consistent with previous cost-effectiveness studies
of smoking ces ation strategies (12,13), we did not include
medical costs incurred by the survivors in subsequent years .
It is proper, nevertheless, to consider the additional medical
costs incurred for routine care during the years of life gained
and, therefore, we included those costs that were incurred
subsequently in a sensitivity analysis .
We did not include indirect costs to the patients, such as
the time lost from work while participating in the program
because the initial contact occurs while the patient is hospitalized, and subsequent contacts are very brief . We also did
not consider the effect of smoking cessation itself on decreasing days missed from work as a result of illness averted .
Sensitivity analysis. Each variable was examined over a
wide range in sensitivity analyses . The effectiveness of the
program, measured as the number of additional quitters/100
smokersenroledintheprogam,variedfrom0to3
.
quit ers/10 smokers (baseline as umption 26/10 smokers)
Discounted life-years gained varied from 0 .1to5(baselin
assumption 1 .7) . If we assume that there was no survival
benefit to the quitters after the 8 years for which Aberg et al .
(5) fol owed up their cohort, then the discounted life-years
saved is 0 .3 (calculated as before up to i = 8), wel within the
range in the sensitivity analysis . The costs of the program
varied from $50 to $2,0 0/patient (baseline as umption $10 )
.
Annual discounted medical costs incurred by survivors
varied from $0 to $20,000 . This range encompasses a recent
estimate of $2,100 for the annual cost of medical therapy for
.
patients after myocardial infarction (14)
ASKS
Cost-effectiveness . The cost of the recently published
smoking cessation program compared with usual care for
patients who have had an acute myocardial infarction was
estimated to be $380 for each smoker who quits (the product
of the cost per participant [$100] and the number of enrollees

Downloaded From: https://onlinejacc.org/ on 08/17/2014

16 9

4
6
him After Mi

4
6

YearsAly91

ne de to produce an ex-smoker 13,81)

. On the basis of the
preceding analysis, each person who stopped smoking after
acute myocardial infarction gained approximately 1 .7years
compared with patients who continued to smoke . Therefore,
the incremental cost-effectiveness is estimated to be approximately $220 for each adcitional year of life saved (the
quotient of the cost of the program for each ex-smoker
produced [$380] and the number of years gained by each
ex-smoker [1 .7]) .
Sensitivity analyses . The cost/year of life saved is calculated to be <$6,0 0 if the program helps at least I ad itional
smoker to quit for every 100 smokers enrolled (Table 2) . The
program has a cost-effectiveness ratio of <$20,000/year of
life saved as long as at least three additional smokers of

Table 2 . One-Way Sensitivity Analyses

Baseline
Effectiveness of program in decreasing smoking
(baselin2610)
3/10
6/10
1/100
31A8)0
Discounted life-years gained
(baseline 1 .7)
5
3
1
0.10
Cost of the smoking cessation program
(baseline $100)
$50
$1,000
$2,000
Medical care costs incurred during years gained
(baseline $0)
$2,100
$5,0
$10,000

Cost-Effectiveness
($/year of life saved)
220
Ito
980
5180
19,60
80
130
380
3,850
110
2 .260
4,520
2,30
5 .230
10,23

KRUMHOLZ ET AL .
SMOKING CESSATION AFTER MYOCARDIAL INFARCTION

170

1000

JACC Vol . 22, No .6

:1607-2

Table 3 . Comparison of Cost-Effectiveness of Interventions for

Patients With Myocardial Infarction

Base Case
he,

Novembr15,93

Cost-Effectiveness
Per Year of Life
Saved ($)

30
2000
Coat of Program, $

40

50

Viprt S . Two-way sensitivity analysis in which the cost and

effectiveness of the program are varied, Lines represent coordinates

that would result in a cost-effectiveness ratio of $20 .000/year of life

saved (YOLS) and $10,0001year of life sued . Coordinates that are
above and to the left represent more favorable ratios, The baseline

result is also indicated .

every 1,000 patients enrolled are able to quit . Even if the
effectiveness of the program wits estimated to be less than
the baseline analysis, the program remained extremely costeffective . If the discounted years gained from smoking
cessation were assumed to be as low as 0 .1, the costef ectivenes of the program stil did not exce d $4,0 0/year
of life saved . Furthermore, given the baseline assumptions
for effectiveness, the cost of the program could be as much
as $8,840/participant and stil remain <$20,0 0/year of life
saved .
The cost of the program and its effectiveness in producing
ex-smokers varied together in a two-way sensitivity analysis
(Fig .3)
. The program could cost much more than our
baseline estimate and still be relatively cost-effective, even if
the program were less effective than assumed, For instance,
even if the cost of the program were as high as $2,010
Participant, the cost-effectiveness of the program would be
<$10,
year of life saved so long as it could help 12
additional smokers to quit for every 100 enrolled .

Discussion
The principal finding of our analysis is that a smoking
cessation program for survivors of an acute myocardial
infarction is an extremely cost-effective intervention, Our
baseline analysis was modeled on a recently described
nurse-managed program that increased smoking cessation
ratesby2610 smokers
. We showed that this program is
much more cost-effective than the use of either boo
adrenergic antagonists after myocardial infarction in high
riskpaten($4,7
yearofliesavdin19 olars)(15or
coronary artery bypass grafting for left main coronary artery
ster osis in patients with severe angina ($7,0 0/qualityadjusteyaroflien19dolars)(16,
two interventions
that are considered very cost-effective by current standards .
Sensitivity andbyUses . Although our baseline calculations
were based on a successful nurse-managed smoking cessa-

Downloaded From: https://onlinejacc.org/ on 08/17/2014

Postinfarction
Smoking cessation program
Beta-adrenoceptor antagonist therapy
(Goldman et al .[15)
Low risk patient
Medium risk patient
High risk patient
pwal0rction
Thrombolytic therapy (Krumholz et al .14)
70yearsold
75yearsold
80 years old
Coronary care unit ifineberg et al .17)
ProbailtyofMI5%
Probability of MI 201N)

231

27,0)
7,40
4,70

21,200
22AX)
21,60)
260,1
60,

MI * myocardial infarction .

Lion program, our results suggest that even smoking cessation programs with much lower effectiveness and higher cost
would have a cost-effectiveness that is superior to that of
many other interventions in cardiovascular medicine (Table
3) . If the program were as effective as that described by
Taylor et al . (4),
yielding a decrease in smoking of 26/10
smokers, then it could cost as much as $2,000 for each
participant and still have a cost-effectiveness ratio that is
< We that of the hospital stay in a coronary care unit for a
patient with a 201% probability of having a myocardial
infarction, a strategy that has been estimated to cost approximately$60, /yearofliesavdin19 olars(17)
. If the
program cost only $100/smoker, as in our baseline analysis,
then an increase in the quitting rate by only I person for each
169 smokers enrol ed would result in a cost-ef ctiven s
ratio <$10,000/year of life saved .
Assumptions of the study. Throughout the analysis, when
information was equivocal, we made assumptions that
tended to bias the analysis against the intervention . For
instance, despite the likelihood that smoking cessation
would result in savings from an expected decrease in the
incidence of cancers, strokes and respiratory-related diseases, we did not include this factor in the model . The
inclusion of estimates of these savings would have resulted
in an even more favorable cost-effectiveness ratio .
Our assumptions about reinfarction after smoking cessation were also conservative .The190U
.S . Surgeon General's Report (2) concluded that smoking cessation after acute
myocardial infarction decreases the rate of reinfarction .
However, because not all studies have found a statistically
significant reduction in reinfarction rate, our analysis assumed that smoking cessation had no effect on reinfarction
rates . The cost-effectiveness ratio of the program may be
even more favorable if it were assumed that smoking cessation prevents reinfarction .

JACC Vol. 22, No .6

November 1 .5 .193
1:697-02

Despite our conservative assumptions, our conclusions

emphasize that resources devoted to smoking cessation
programs after myocardial infarction are well spent . Furthermore, our analysis extends the observations from the previous reports . Other investigators have analyzed the costeffectiveness of physician counseling and of nicotine gum as
an adjunct to physician counseling against smoking (12,13,18)
.
Those analyses reported that the cost-effectiveness of physician counseling was <$2,000/year of life saved and that nicotine gum as an adjunct to physician counseling was <$10,00/
year of life saved . However, neither of those analyses
considered secondary prevention after myocardial inia,.-ction .
Limitations of the study. There are some important limitations to our analysis . Because no randomized trial of
smoking cessation has been published, we based our estimate of the effectiveness of smoking cessation on the large
observational studies (5-10)
. In those studies, the two comparison groups were often very different :Thesmokerswho
quit commonly had a worse risk profile (e .g ., more severe
left ventricular failure, higher peak creatine kinase enzyme
levels) than that of the smokers who continued . Therefore, it
is more likely that the observed relative reduction in mortality at 5 years was the result of smoking ces ation rather
thanselction(5)
.
The baseline estimate of the discounted survival benefit
was calculated using a single declining exponential as a
model of survival in both groups . This approach has been
validated previously as an appropriate technique in situations where mortality is dominated by a single disease
process (11) . Figure 2 shows that this fits the actual data well
for at least the 1st 8 years after myocardial infarction . This
method may overestimate the survival benefit in the long run
because there are no data to show that the survival benefit to
quitters persists over a long peiiod and, over time, the
mortality rate in both groups is likely to increase rather than
remain constant as in the single declining exponential model .
Nevertheless, discounting future life-years minimizes any
discrepancy because it gives greater weight to the earlier
years . Furthermore, the range of the sensitivity analysis for
discounted life-years saved included values below that calculated, assuming that there was no survival benefit after
8 years .
We did not model a relapse rate after smoking cessation
because smoking status in the published studies was evaluated early after the acute myocardial infarction and not
for
subsequently verified . In the study by Aberg et al . (5),
instance, smoking ces ation was as es ed only 3 months
after acute infarction . Many of those patients may have
subsequently had a relapse ; thus, the relapse rate is incorporated into the published survival curve .
We did not model costs incurred during years gained in
the baseline analysis because previous studies of smoking
cessation have not included those costs . Those costs are
difficult to estimate because costs incurred by years gained
may be offset by savings from the prevention of costly
morbidity that is averted as a result of smoking cessation .

Downloaded From: https://onlinejacc.org/ on 08/17/2014

KRUMHOLZ ET AL .
SMOKING CESSATION AFFER MYOCARDIAL INFARCTION

170

.1

Large observational studies with careful cost accounting

would be necessary to document the effect of smoking
cessation on subsequent medical costs . Nevertheless, the
sensitivity analysis did include a wide range of costs and
showed that the program remained very cost-effective even
if annual medical costs were as high as $10,000 .
The lack of data about the benefit of smoking cessation
after acute myocardial infarction for women and older
patients made it difficult to model explicitly the effect of age
and gender on the cost-effectiveness of this intervention .
Many of our assumptions are based on studies of middleaged men and women . With a paucity of empiric data it is
difficult to model confidently how age would influence many
of the assumptions in this model . However, there is evidence
that older smokers do benefit from smoking cessation (19)
.
In our analysis we sought to determine whether this intervention would be cost-effective over a wide range of assumptions that would encompass reasonable estimates pertaining
to men and women of any age .
Finally, our analysis did not consider the potential benefit
of transdennal nicotine patches . Recent studies have elemonstrated the value of these patches in smoking cessation
programs (20,21) . However, there are no published trials of
the patch in patients after myocardial infarction . Furthermore, there are theoretic concerns about the hazards of
administering nicotine to patients after an acute myocardial
infarction . Nevertheless, if the patches are safe and effective
in this setting, our results suggest that the benefits of
increasing smoking cessation with transdermal nicotine therapy could ea , ,ily outweigh its costs . Further studies of the
role of transuermal nicotine patches in this population are
needed .
Conclusions . We believe that our analysis has important
implications for the approach to the treatment of smokers
with acute myocardial infarction . Taylor et al .(4)demonstrated that this intervention is effective after myocardial
infarction, and we showed that it is more cost-effective than
commonly accepted treatments, such as beta-adrenergic
antagonist therapy . Organized smoking cessation programs
with telephone follow-up by trained professionals should be
a part of the care for every smoker who suffers a myocardial
infarction .
We thank Stephen G . Pauker. MD for helpful discussions of the manuscript .

References
I . U .S . Public Health Service . Smoking and health . Report of the Advisory
Committee to the Surgeon General of the Public Health Service . Washington . DC :U .S . Department of Health . Education and Welfare . Public
Health Service, Centers for Disease Control, PH$ publication No .103,
1964 .
2 . U .S . Department of Health and Human Services, The Health benefits of
smoking cessation . Washington. DC :U .S . Department of Health and
Human Services, Public Health Service . Centers for Disease Control,
Center for Chronic Disease Prevention . Office of Smoking and Health .
. M0.
DHHS Publication No .(CD)90-8416

4 2
17 U

KRUMHOLZ ET AL.
SMOKING CESSATION ALTER MYOCARDIAL INFARCTION

3 . Siegel D, Grady D, Browner WS, Hulley SB . Risk factor modification

;109
:213-8
.
after myocardial infarction . Ann Intern Met" 1
4 . Taylor AB, Houston-Miller N, Killen JD, DeBusk RE Smoking cessation
.
after acute myocardial infarction: ef ect of a nurse-man ged intervention
;13
:18-23
.
AnIterMd190
5 . Aberg A, Bergstrand R, Johansson S, et al . Cessation of smoking after
; after 10 years . Br Heart J
myocardial infarction:ef ctsonmrtai
1
;49 :416-2
.
6 . Johansson S, Bergstrand R, Pennert K, et al . Cessation of smoking after
.
myocardial infarction in women : ef ects on mortality and reinfarctions
AmJEpidemol1985
;121 :823-1
.
7 . Perkins J, Dick TBS . Smoking and myocardial infarction :secondary
.
prev ntion, Postgraduate Med J 1 5
;61 :295-30
8, Sparrow D, Dawber TR, Colton T . The influence of cigarette smoking on
osis after a first myocardial infarction :areportfromtheFraminghamStudy,JChronicDs1978
:31 :4532
.
9, Hedback B, Park J, Five-year results of a comprehensive rehabilitation
.
prog am e after myocardial infarction, Ear Heart J 1987
;8:23410. Daly LE, Mulcahy R, Graham 1M, Hickey N, Long term effect on
mortality of stopping smoking after unstable angina and myocardial
infarction .BMJ1983
:287
:324-6
.
. A convenient approximation of life
1 , Beck JR, Kas ire JP, Pauker 86
expectancy (the "DFALE"), 1 . Validation of the method, Am J Mod
19823
:8597
.
12, Oster G, Huse DM, Delea TE, Colditz GA . Cost .effcctiveness of nicotine

Downloaded From: https://onlinejacc.org/ on 08/17/2014

Novembr15,93

JACC Vol . 22, No.6

:1697-02

gum as an adjunct to physician's advice against cigarette smoking . JAMA

1986
;256
:135-8
.
13 . Cummings SR, Rubin SM, Oster G . The cost-effectiveness of counseling
smokers to quit .JAM198
;261
:75-9
.
14 . Krumholz HM, Pasternak RC, Weinstein MC, et al . Cost effectiveness of
thrombolytic therapy with streptokinase in elderly patients with suspected acute myocardial infarction .NEnglJMed192
;327
:7-13
.
15 . Goldman L, Sia STB, Cook EF, Rutherford JD, Weinstein MC . Costs and
effectiveness of routine therapy with long-term beta adrenergic antagonists after acute myocardial infarction .NEnglJMed198
;319
:152-7
.
16 . Weinstein MC, Stason WB. Cost-effectiveness of coronary artery bypass
surgery .Circulation1982
;6Supl
:il-56
.
17 . Fineberg HV, Scadden D, Goldman L . Management of patients with a
low probability of acute myocardial infarction :cost-ef ctivens of
alternatives to coronary care unit admission, N Engl J Mod 1
;310
:
1 .301-7
.
18 . Tsevat J. Impact and cost-effectiveness of smoking interventions . Am J
Med192
;93SupilA
:43Sm7
.
19, LaCroix AZ, Lang J, Scher P, et at
. Smoking and mortality among older
men a d women i thre com unit es, N EngI J Mod 19
;324
:16925
.
20 . Tonnesen P, Norregaard J, Simonsen K, Sawe U . A double-blind trial of
a 16-hour transdermal nicotine patch in smoking ces ation
. N Fngl J Med
19
;325
:315
.
21 . Transdermal Nicotine Study Group . Tt'ansdermai nicotine for smoking
cessation .JAM19
;26
:31.8 .