C1GARETTE SMOKING AND HEALTH

AMERICAN

THORACIC

SOCIETY. MEDICAL SECTION OF AMERICAN

THIS OFFICIAL STATEMENT OF T(.ll:. AMERICAN

OF DIRECTORS.

NOVEMBER

THORACIC

LUNG ASSOCIATION

SOCIETY WAS ADOPTED BY THE

ATS

BOARD

1984

Purpose of the Statement

In 1964, the first Surgeon General's Report
on Smoking and Health stated that cigarette
smoking was a causal risk factor for lung cancer and a contributing factor for other diseases. Over the next twenty years, the evidence
linking smoking and many diseases became
increasingly certain so that the 1983 Surgeon
General's Report on Smoking and Health concluded that cigarette smoking is the largest
avoidable cause of deat h and disability in the
United States. Because the adverse effects of
cigarette smoking continue to be major health
problems, a committee of the Scientific Assembly on Environmental and Occupational
Health of the American Thoracic Society has
prepared this statement on cigarette smoking
and health.
This statement summarizes the adverse
health effects of cigarette smoking, addresses
the addictive nature of the habit, and reviews
aspects of smoking cessation and prevention.
Emphasis is placed on information that has
accumulated since the 1964 Surgeon General's
Report. We recognize that there are many nonpulmonary health effects of cigarette smoking and discuss some of them in this report,
but we will address adverse pulmonary health
effects in more detail. We are stating a consensus point of view and not an in-depth review and will provide only a limited number
of references because of space limitations and
the availability of detailed documents such
as the Surgeon General Reports.
Because we conclude that smoking causes
many adverse effects, we need to state how
a judgment of a causal relationship
is made.
Extensive evidence has been published on the
effects of cigarette smoking and health. The
evidence includes laboratory investigations of
smoke characteristics and biological activity,
short-term clinical investigations, and epidemiological studies that examine the prevalence
of disease in populations in relation to cigarette use Direct experiments with humans,
regarded by some as scientifically necessary
for establishing causality, are neither ethical
nor feasible and have not been considered
necessary for many other deleterious agents
such as ionizing radiation and asbestos. The
judgments concerning causality of associations between tobacco smoking and diseases
stated in this document are based on all available evidence, both from the laboratory and
from human populations. Associations have
been judged as causal on the basis of biological plausibility, reproducibility, an appropriate temporal relationship between exposure

and disease, and sufficient strength of the association with evidence of a dose-response
relationship. Specificity of the association of
exposure and disease strengthens the argument for causality.
Prevalence of Cigarette Smoking
Although the prevalence of cigarette smoking has decreased since the publication of the
first Surgeon General's Report on Smoking
and Health in 1964, millions of Americans
still smoke. In 1965, 52070 of men and 34070
of women over age 20 were cigarette smokers.
By 1980, these percentages had decreased to
380/0 for men and 30070for women. A Gallup
survey in 1984 indicated that the rate of smoking among adults declined to 29070. The per
capita consumption of cigarettes in the United
States also decreased from 4,345 cigarettes per
adult over 18 in 1963 to 3,494 in J983, a 20070
decline over the 20-year period. Smoking
among younger individuals, especially young
women, remains a major public health concern. Several studies indicate that smoking
is now more prevalent among teenage women
than men. In spite of the recognized adverse
health effects of cigarette smoking, about
15010of adolescents (age 12 through 17) currently smoke cigarettes.
Cost of Smoking
The enormous economic COStSof smoking
must be considered in addition to the adverse
health effects and the concomitant reduction
in quality of life. In terms of health care, the
COStsattributable to cigarette smoking exceed
$17 billion per year. Wben lost work and
productivity are added to direct medical costs,
the tOLaICOStto society is estimated to exceed
$41 billion per year or $180 per capita. If these
costs were borne by smokers in the form of
cigarette taxes, the price of cigarettes would
rise ro over $3 per pack. For the individual
under 50 who smokes over 2 packs per day,
thesumof lifetime loss in earnings and medical expenses is estimated to exceed $34,000
How Cigarettes Work
The burning cigarette is a chemical factory
that generates thousands of different compounds. The precise chemical composition of
smoke depends on the type of cigarette and
the way in which it is smoked. Major toxic
constituents of cigarette smoke include but
are not limited to carbon monoxide, nicotine,
and particulates that contain most of the carcinogenic polynuclear aromatic hydro car-

bons. The addlctive properties of nicotine and

the myriad of chemical compounds in tobacco
smoke make it unlikely that a "safe" cigarette
can be produced. The low tar and low nicotine cigarettes now being marketed
may be
marginally safer in terms of lung cancer than
their predecessors on a unit basis. However,
there is concern that Smokers of these products adjust their consumption
and smoking
pattern to achieve the nicotine blood concentrations reached with the regular types of cigarettes and thus inhale more total smoke.
The pathophysiologic mechanisms underlying the bealth effects of cigarette smoking
are complex because of the myriad smoke
components and their many direct and indirect interactions with environmental and
genetic factors. Because cigarette smoking is
known to be an independent risk factor for
chronic obstructive pulmonary disease, lung
cancer, and coronary heart disease, patients
with other recognized genetic, metabolic, or
occupational risk factors for these diseases
must be strongly advised not to smoke. Unfortunately, we do not know why some
smokers develop severe chronic obstructive
pulmonary disease, lung cancer, or coronary
heart disease, whereas other individuals are
minimally affected. We do know that, as a
group, smokers have a higher risk of developing chronic obstructive lung disease, lung cancer, and coronary heart disease (ban those
who do not smoke, Currently, there is no way
of predicting which individuals will develop
severe disease and which individuals will not.
Disease can be most effectively prevented by
avoiding smoking all together.
General Health Effects and Mortality
Since the first Surgeon General's Report was
published in 1964, compelling epidemiologic,
clinical, and experimental evidence has accumulated to strengthen the association between cigarette smoking and both mortality
and morbidity from a wide range of diseases.
The overall mortality ratio for adult cigarette
smokers versus adult nonsmokers is about 1.7.
The mortality ratio is the ratio of the number
of observed deaths in smokers to the number
expected from nonsmokers. The mortality ratio in smokers increases with the amount
smoked and is directly proportional
to the
duration of cigarette smoking. Mortality rauos are also higher for those who start smoking at younger ages. While it is hard to calcuRerrlntcd

from

AMERlCAN (tI;VIEW Of RESPIRATORY DISEAse.

Vo. 132. No. S. November t9SS.pp, 1L33-1I36

AMERICAH TlfOfUCIC

SOCIETY

late exactly the effects of smoking on life ex- can be detected with very sensitive tests in
pectancy. all estimates lead to the conclusion
many cigarette smokers after 10 to 15 years
that life expectancy at any age is significantly
of smoking. However, small airways disease
shortened by cigarette smoking. For examis not necessarily a forerunner of the severe
ple, a 30- to 35-year-old, two-pack-a-day
functional impairment seen with emphysema.
smoker has a life expectancy 8 to 9 years The l-second forced expiratory volume, the
shorter than a nonsmoker of the same age.
The excess mortality noted in smokers is FEV" is the most predictive and reproducible pulmonary function test for both epidemigreatest for the 45- to 54-year-old age groups
ologic and clinical studies. In nonsmokers,
for both men and women. Cigarette smoking
the FEV I declines with age during adult life
is the largest preventable cause of premature
at the rate of about 20 10 30 m1 per year. In
death in the United States today.
most smokers, the rate of decline is increased
In addition to its effect on mortality,
to about 30 to 45 ml per year. In the 10 to
ciga- rette smoking causes substamlal
150;0of smokers who develop clinically sigmorbidity. Both men and women who smoke
nificant
impairment, the rate of decline is
report more acute and chronic symptoms
about 80 to 100 ml per year. The two puland illnesses than people who have never
monary function characteristics, which are
smoked.
helpful in identifying the smoker who is
likely
General Pulmonary Effects Cigarette
to develop severe pulmonary impairment, are
smoking produces structural and functional
a relatively low FEV, by middle-age and a
changes in both the conducting airfaster than expected fall in FEV, from year
ways and the pulmonary parenchyma. The to year. Patients with emphysema associated
structural changes in the large airways con- with smoking cigarettes have a reduced single breath di ffusing capacity, but a low diffussist of hypertrophy and hyperplasia of the
ing capacity is not specific for emphysema.
mu- cous glands. These changes are
responsible for the increase in mucus
lung cancer
production that leads to the increased
cough and sputum production. Structural
Of the many adverse consequences of cigachanges in smaller air- ways range from
rette smoking, lung cancer was the first to be
relatively mild inflamma- tion to narrowing
causally linked to tobacco smoke exposure.
and closure of airways due to inflammation,
The 1964 Surgeon General's report concluded
goblet ceU hyperplasia, and intraluminal
that cigarette smoking was causally related
mucus. Changes in the paren- chyma
10 lung cancer in men and was probably of
include increased numbers of inflam- matory similar importance in women. Since that recells and ultimately destruction of the
view, abundant additional data have conalveolar walls. most commonly in the central
firmed that cigarette smoking is the major
part of the lobule, and therefore termed cen- cause of lung cancer in the United States in
trilobular emphysema.
both men and women.
Airway disease attributable to cigareue
The risk for individual smokers varies
smoking without coexisting emphysema is not with smoking practices. Dose-response
usually associated with severe impairment of relation- ships have been demonstrated with
pulmonary function. Smokers with severe duration and amount of smoking. Deeper
functional impairment usually have an apinhalation and earlier age of starting increase
preciable amount of emphysema. Tbus, it is lung can- cer risk. In some studies,
likely that the 10 (0150/0 ofcigareue smokers modifications of the cigarette to reduce the
who develop appreciable impairment of their yield of tar have been associated with
lung function are the ones who have devel- modest reductions of lung cancer risk.
oped emphysema in addition to bronchitis.
However, the current low tar products have
Although an explanation for the failure of
not been evaluated in epi- demiologic
all smokers to develop emphysema remains
studies.
to be established, new hypotheses on the
In the U nired States, lung cancer
pathogenesis of emphysema provide insight. incidence and mortalityhave been lower in
In the lung, there is a balance between facwomen than in men. The difference reflects
tors causing proteolysis and factors protecttemporal trends of smoking in the two
ing the lung from proteolysis. Neutrophil
groups. Cigarette use became widespread
elastase Is thought to be responsible for pro- among meo early in this century, whereas
teolysis, whereas alpha-I-anti protease is con- large numbers of women did not smoke until
sidered to be the major protective factor. Cig- the 1940s. As a result, rbc pattern of
areue smoke can both increase the influx of
increasing lung cancer in women has
neutrophils, and thereby the burden of neuparalleled that seen earlier in men but has
trophil elastase, and inactivate alpha-l-andlagged by about 2S years. In 1985, lung canprotease, Thus, smoking cigarettes increases cer will become the leading cause of cancer
the potential for proteolysis. However, the bio- deaths in women, a preeminence Ihat is
logic iI1teractions bet ween cigarette smoke and largely attributable to cigareue smoking.
the lunS are undoubtedly much more complex and further research is needed in this area.
Other Cancers
The structural changes associated with
Malignancies other than lung cancer bave also
cig- areue smoking are associated with
been linked to cigarette smoking. Numerous
functional impairment. Mild functional
reports of the Surgeon General have applied
impairment, which is almost certainly
the criteria for causality on a site-speciflc bacaused by disease at the level of the small
sis. In the 1982 Surgeon General's report on
peripheral airways.
cancer, cigarette smoking was designated as
2

a "major cause" for cancer of the lung, larnyx, oral cavity, and esophagus in the United

Slates. Cigarette smoking was also identified

as a "contributory factor" for cancers or the
bladder, kidney, and pancreas. Reduction of
smoking should have direct and dernonstrable benefits for those sites where cigarette
smoking is a major cause.
Cardiovascular Effects
Cigarette smokers experience a 700/0 greater
death rate due to coronary heart disease than
nonsmokers. The first Surgeon General's Report concluded that there was an association
bet ween smok ing and coronary heart disease,
but that evidence was insufficient [0 make
the judgment of a causal relationship. By
1979, however, it was recognized that cigarene smoking was a major risk factor for coronary heart disease among both men and
women in the United States, In 1980. cardiovascular disease accounted for nearly half of
all U.S. deaths (960,000 of 1,980,000 total
deaths), and nearly 60% of these deaths were
due to coronary heart disease. Investigators
estimate that 30 to 4001. of deaths due to coronary heart disease are auributable to cigarette
smoking.
Pregnancy and Smoking
Cigarette smoking by pregnant women leads
to a variety of adverse consequences for the
unborn child. The mortality rate of babies
of mothers who smoke is higher t han the mortality rate of babies whose mothers do not
smoke. Babies born to women who smoke
during pregnancy are 200 grams lighter on
the average than babies born to nonsmokers.
The more a woman smokes, the greater the
reduction in the birth weight of her baby. Because the duration of fetal gestation does not
appear to be sbonened by maternal cigarette
smoking, the lower birth weights seen in babies of smokers appear to be due to retardation of feral growth.Jn multiple studies, pregnant women who smoke have a 10 to 20010
increased risk of spontaneous abortion. This
correlation continues to hold when other risk
(actors - age, panty, education, race, and socioeconomic status-are controlled. Other
complications of pregnancy that are seen signi ficantly more often in smoking mothers include placenta previa, abruptio placenta
bleeding during pregnancy, and prematur~
rupture of membranes. These conditions also
contribute 10 increased fetal death. Those who
provide prenatal care should counsel pregnant
women about the adverse effects of smoking
on pregnancy and infant health.
OccupatIon and SmOking
Smoking is a major risk factor in certain occupational lung diseases. Airways obstruction
is the most common occupationally associated class of lung disease in which smoking plays a major role. Smoking and occupational exposure to coal mine dust or grain or
couon dust appear to act in an additive fashion to produce bronchitis and airways obstruc-

tion. Clinically severe emphysema is mainly

linked to cigarette smoking. Cigarette smoking plays an important etiological role in occupationally associated bronchogenic cancer.
For two exposures, asbestos and radon daughters, the evidence supports synergism with cigarette smoking in the development of lung
cancer. Nonsmoking
asbestos insulation
workers have a fivefold increased risk for lung
cancer compared to nonsmoking individuals
not exposed to asbestos, but smoking asbestos
insulation
workers have a greater than
fiftyfold increased risk for lung cancer. Hence,
smoking is a very powerful risk factor that
increases the risks of air-flow obstruction and
lung cancer associated with certain occupational exposures. Therefore, smoking cessation programs in those at-risk industries
could have an important impact on the prevention of smoking related occupational lung
diseases.
Passive Smoking
Passive smoking refers to the involuntary exposure of nonsmokers,
both children and
adults, to tobacco combustion
products. In
enclosed spaces, smoke accumulates
and the
concentration
varies with the number of
smokers, with the type of smoking, and with
the characteristics of the room, especially the
ventilation. Although the exposures from active and passive smoking are not identical,

year smoker of 1 pack a day has inhaled over

I million times, and this repetitive act creates
many strong cues for smoking such as coffee,
alcoholic beverages, finishing a meal, driving, and other activities. Smoking by family
members and friends is strongly associated
with starting in adolescence; smoking by
spouses and friends is strongly associated with
difficulty in cessation during adulthood.
While it is tempting to try to identify some
crucial motive for smoking, research indicates
that there is no single factor. Although people may begin smoking for only one of the
reasons mentioned above, they usually maintain the habit because of most or all of them.
Smoking becomes a part of many activities,
and these associations help explain the difficulty of quitting and likelihood of relapsing.
The Role of Addiction
The definition of addiction is not simple. Current criteria for defining a substance as addictive or leading to dependence include the
development of dependence and compulsive
use; psychoactive, behavioral, and reinforcing effects; difficulty in quitting and likelihood of relapsing; and withdrawal symptoms
following cessation.
Measured
by these
criteria, smoking is now generally accepted
as being addictive and nicotine appears to be
a major reason for the addiction.
It is important to understand that addic-

The hardest part of smoking cessation is

not quitting but staying off cigarettes. Although initial rates of quitting are high, many
resume smoking. In most programs between
20 and 30070 of former smokers are free from
cigarettes after I year. A program that is only
20 to 30% successful may not sound impressive. However, two factors suggest these success rates are significant for public health programs. First, the tremendous economic and
health costs of smoking make any success important. Second, those who permanently quit
frequently report having stopped and relapsed
an average of 2 to 3 times before they succeeded. Relapses appear to be caused by exposure to the smoking of others; by anger,
anxiety, and sadness; and by heightened
desires for cigarettes due to withdrawal.
Smokers may derive benefit from programs
that offer support or continued treatment for
several weeks to several months after quitting. Noting that most relapse before they succeed, it is helpful to avoid an "all-or-nothing"
approach to relapse. Relapsing once does not
mean a case is hopeless.
Physicians should take a major role in
promoting smoking cessation. Simply counseling each smoker for several minutes has
been shown to help as many as 5% to quit
for at least I year. Although this percentage
sounds
insubstantial,
the frequency
of
physician-patient encounters, the relative cost

AMERICAN

due to cardiovascular disease within the first

year of stopping. As the duration of smoking cessation lengthens, the former smoker's
risk approaches that of the never smoker. The
cigarette-related respiratory symptoms of
cough and sputum production usually improve rapidly after quitting. Unfortunately,
the smoker with significant chronic air-flow
obstruction cannot anticipate major functional improvement after cessation. However,
the rate of further deterioration of lung functionalmost always decreases. The relative risk
of lung cancer for the ex-smoker decreases
as the cigarette-free period increases and approaches that for nonsmokers after 15 to 20
years. Smoking cessation is also beneficial for
decreasing the risk of cancers of the larnyx,
oral cavity, esophagus, and urinary bladder.

THORACIC

dren and may increase the risk of lung cancer

in the nonsmoker. Smoking cessation is difficult because of nicotine addiction and psychological and social factors. Physicians and
other health professions must be active both
in helping people quit smoking and in preventing nonsmokers from starting. We recornmend a national effort focused on preventing young people from starting to smoke.
This statement was prepared by a subcommittee of the Scientific Assembly on Environmental and Occupational Health. The members of this committee are:
ROBERT J. MASON, M.D.,

A.
EDwrN

Smoking Prevention

SONIA

B.

Chairman

BUIST, M.D.
FISHER, PH.D.

JAMES A. MERCHANT,
JONATJ:iAN

The future lies with smoking prevention. Raising a generation of nonsmoking Americans
must be a goal of all health professionals. As
with smoking cessation, simply acknowledging the risks of smoking may not be sufficient in and of itself to help youngsters avoid
tile habit. Experience in several European
countries indicates that substantial reductions
in adolescent smoking occur when strong

SOCIETY

CAROLYN

M. SAMET,

H.

WE,LSli,

M.D.
M.D.
M.D.

Bibliography
I. Correa P, Pickle LW, Fontham E. Lin Y,
Haenszel W. Passive smoking and lung cancer.
Lancet 1983; 2:595-7.
2. Doll R. The smoking-induced
epidemic. Can
J Public Health 1981; 72:372-81.
3. Doll R, Pew R. The causes of cancer: quantitative estimates of avoidable risks of cancer in

vironmental
Health Hazards. Assembly of Life
Sciences. Washington.
DC: National Academy
Press, 1981.
12. Oster G, Graham A. Colditz A, Kelly Nl.. The
economic costs of smoking and the benefits of quitting. Lexington, MA: lexington
Books. 1984.
13. Royal College of Physicians.
Smoking or
health. A Report of the Royal College of Physicians. London: Beekman Pubs .. 1977.
14. Russell MAH, WilsonC, TaylorC. Baker CD.
Effect of general practitioners' advice against srnoking. Brit Med J 1979; 2:231-5.
15. Sexton M, Hebel JR. A clinical trial of change
in maternal smoking and its effect on birth weight.
JAMA 1984; 251:911-5.
16. Trichopoulos D, Kalandidi A, Sparros L. MacMahon B. Lung cancer and passive smoking. Int
J Cancer 1981; 27:1-4.
17. U.S. Department of Health and Human Services. Public Health Service. The health consequences of smoking -chronic obstructive lung diseasea report of the Surgeon General. Washington, DC:
U.S. Government
Printing Office, 1984 (DHEW
Publication
No. (PHS) 84-50205).
18. U.S. Department of Health and Human Services, Public Health Service. The health consequences of smoking - cancer - a report of the Surgeon
General. Washington, DC: U.S. Government Printing Office, 1982 (DHHS Publication
No. (PHS)
82-50179).
19. U.S. Department of Health and Human Ser-

vices, Public Health Service, Surgeon General's Ad-