have been published for the 12-year period for women; the main
body of published data for women is based on the first 4-year
period of the follow-up.

THE SWEDISH STUDY (1)

A national probability sample of 55,000 Swedish men and
women was surveyed in 1963, by a mailed questionnaire to
which 89 percent of the sample responded. Information was col-
lected on smoking status at the time of the query and at
specified intervals during the previous 9 years according to type
and amount of smoking and degree of inhalation. The question-
naire identified age, sex, location (urban, nonurban), income,
and occuption of each subject. A 10-year follow-up on smoking-
related mortality was published in 1975.

THE CANADIAN VETERANS STUDY (4)

Beginning in 1955, the Department of National Health and
Welfare, Canada, enrolled 78,000 men (veterans on pension) and
14,000 women (mostly widows of veterans) in a study of
smoking-related mortality. Information was obtained on age,
detailed smoking history, residence, and occupation. During the
6 years of follow-up, 9,491 of the men and 1,794 of the women
died. No recent follow-up has been reported.

JAPANESE STUDY OF 29 HEALTH DISTRICTS (7)

In late 1965, a total of 265,118 men and women in 29 health
districts in Japan were enrolled in a prospective study. This
represented from 91 to 99 percent of the population aged 40 and
older in these districts. This study provides a unique opportu-
nity to examine the relationship of cigarette smoking to death
rates in a population with genetic, dietary, and other cultural
differences from previously examined Western populations. At
the time of the eighth year of follow-up 11,858 deaths had oc-
curred and there were 1,269,382 person-years of observation.
For women, however, the main body of published data is based
on 5 years of follow-up.

THE BRITISH DOCTORS STUDY (2)

In 1951, the British Medical Association forwarded to all
British doctors a questionnaire about their smoking habits. A
total of 34,400 men and 6,207 women responded. With few excep-
tions, all men who replied in 1951 have been followed for 20
years. Further inquiries about changes in tobacco use and some
additional demographic characteristics of the men were made in

51
1957, 1966, and 1972. More than 10,000 deaths have occurred in
this population during the past 20 years. For women, published
data are available for 11 years of follow-up, and unpublished
data are available for 22 years of follow-up.

THE FRAMINGHAM HEART STUDY (10)

The Framingham Study began in 1948 with a cohort of 2,336
white men and 2,873 white women who were age 29 to 62 at the
beginning of the study and were residents of Framingham,
Massachusetts. Persons were selected by a sample of house-
holds plus enlistment of volunteers. These individuals were re-
called and examined every 2 years thereafter.

The routine cardiovascular examination consisted of a medi-
cal history, physical examination, blood chemistries, body
measurements, vital capacity, chest x-ray and a 12-lead elec-
trocardiogram. Mortality and morbidity were documented in
detail from the routine biennial examination, hospital records,
death certificates, physician records and the next-of-kin.

Information on smoking was obtained at the first examina-
tion (and at several thereafter). A series of monographs and
over 200 articles on the Framingham Study have now become
part of the scientific literature.

Data on the relationship of cigarette smoking to cardiovascu-
lar morbidity and mortality, for both men and women, have
been reported in the Framingham literature, but the longest
reported follow-up period has been 18 years with relatively few
deaths having occurred by then, especially among the women
(11). Data given below are based on a longer follow-up period, 26
years, and have not been published. The study is presently in its
16th biennial cycle.

THE BRITISH-NORWEGIAN MIGRANT STUDY(10)

In October 1962, morbidity questionnaires requesting infor-
mation on personal and demographic characteristics, including
cigarette smoking, as well as symptoms of cardiorespiratory
disease were sent to approximately 32,000 British migrants and
18,000 Norwegian migrants to the United States residing in 12
states. These samples were drawn from the 25 percent random
sample of the entire population for which country of birth was
recorded in the 1960 United States Census. The 12 states in-
volved contained about three-fourths of the British and Norwe-
gian immigrants to the United States. The response rate to the
questionnaire was 86 percent. The respondents were then fol-
lowed for survivorship and cause of death data for 5 years, from
January 1, 1963 through December 31, 1967. The number of

52
morbidity questionnaire respondents and deaths occurring
among them from 1963 to 1967 for ages 45 to 74, by sex, were as
follows.

 

Males Females
Respondents Deaths Respondents Deaths
British 10,103 1,181 9,057 588
Norwegian 5,902 643 5,337 354

Several reports dealing with the prevalence survey and with
a related cross-sectional study of mortality, including data on
cigarette smoking for women as well as for men, have been pub-
lished (13,14,15,16). The main results of the prevalence study —
may be briefly summarized. Four syndromes were considered:
“persistent cough and phlegm,” “chronic bronchitis,” “angina,”
and “possible infarction.” The relation of smoking to the preva-
lence of these symptoms was clearly demonstrated for women
as well as for men. The main results of the cross-sectional mor-
tality study indicated substantial excess mortality for cigarette
smokers, as compared to nonsmokers, for both women and men.

Overall Mortality for Females— Cigarette Smokers Versus
Nonsmokers

MORTALITY RATIOS

In this report the mortality ratio is the basic means of com-
paring cigarette smokers with nonsmokers. It is usually ob-
tained by dividing a “death rate” (or other mortality measure)
for a classification of smokers by the “death rate” (or other
mortality measure) of a comparable group of nonsmokers. The
“death rate” may differ markedly from one study to another. In
Some studies it is calculated by means of person-years and is a
1-year measure; in others it is a probability measure; it may be
a 5-year, 10-year or, as in the Framingham Study, a 26-year
measure. Differences in mortality ratios may arise because of
these factors.

Because of the arithmetic nature of this ratio, there is a tend-
ency for lower ratios to result with higher underlying levels of
mortality. For example, with an underlying mortality level of 10
percent per year for nonsmokers, the mortality ratio for a group
of smokers can at most be 10 if all the smokers died within the
year. With a mortality level of 50 percent for nonsmokers, the
maximum possible ratio is 2. Since “death rates” increase with
age, there is a tendency for the mortality ratios to decline with
age, since its range is restricted.

53
TABLE 2.—Mortality ratios for female cigarette smokers by
number of cigarettes smoked per day and age;
females in 24 states

 

 

 

 

Number of
cigarettes Age Total, 35-84
per day 35-44 45-54 55-64 65-74 75-84 Age-adjusted!
Nonsmokers 1.00 1.00 1.00 1.00 1.00 1.00

1-9 .90 .95 .99 1.09 1.07 .97
10-19 97 1.22 1.31 1.18 1.21 1.19
20-39 1.35 1.54 1.46 1.51 .85 1.45

40+ 1.56 1.96 1.23 1.42 * 1.63
All Smokers 1.12 1.31 1.27 1.31 1.14 1.26

 

‘Adjusted by the direct method using as standard the age distribution of all
women.

*Not shown—less than 5 expected deaths.

SOURCE: Hammond, E.C. (5).

TABLE 3.—Mortality ratios for female cigarette smokers by
number of cigarettes smoked per day and age;
females in the Swedish study

 

 

 

 

Number of

cigarettes Age Total, 18-69
per day 18-39 40-49 50-59 60-69 Age-adjusted
Nonsmokers 1.0 1.0 1.0 1.0 1.0

1-7 1.0 1.6 1.1 9 1.0
8-15 2.3 2.2 1.7 1.4 1.5

16+ 4.5 2.2 1.5 2.2 2.0

All Smokers 1.8 1.9 1.3 1.1 1.2

 

SOURCE: Cederlof, R. (2).

For simplicity, however, mortality ratios are used throughout
this review; it is recognized that these ratios are not strictly
comparable from one study to another nor from one age group
to another.

AMOUNT SMOKED AND AGE

Overall mortality ratios by amount smoked and age are pres-
ented for several of the studies in Tables 2-7. Except for the
Swedish study (Table 3), age-adjusted ratios were calculated for
each level of smoking in each study. Adjustment was by the
direct method, using as standard the age distribution of all

54
TABLE 4.—Mortality ratios for female cigarette smokers by
number of cigarettes smoked per day and age;
females in the Canadian study

 

 

 

 

Number of

cigarettes Age Total, 30+
per day 30-54 55-64 65-74 T5+ Age-adjusted!
Nonsmokers 1.00 1.00 1.00 1.00 1.00

1-9 1.59 1.09 1.05 92 1.20

10+ 2.25 .93 1.20 * 1.43

All Smokers 1.95 1.03 1.10 .95 1.31

 

1Adjusted by the direct method using as standard the age distribution of all
women.

*Not shown—less than 5 expected deaths.

SOURCE: Best, E.W.R. (1).

TABLE 5.—Mortality ratios for female cigarette smokers by
number of cigarettes smoked per day and age;
females in the Framingham Heart Study

 

 

 

 

Number of
cigarettes Age Total, 29-62
per day 29-44 45-54 55-62 Age-Adjusted!
Nonsmokers 1.00 1.00 1.00 1.00
<20 1.42 1.21 1.07 1.30

20 1.84 1.48 1.13 1.62

21+ 2.25 1.14 * 1.72
All Smokers 1.62 1.28 1.07 1.43

 

1Adjusted by the direct method using as standard the age distribution of all
women.

*Not shown—less than 5 expected deaths.

SOURCE: National Heart, Lung, and Blood Institute (10).

women in the particular study. For the Swedish study the age-
adjusted values were taken directly from the report.

Mortality ratios shown in Table 2 are considered especially
important since they are derived from the study with the
largest survivorship experience. Mortality ratios generally rose
with the amount smoked for each age group except for the 75 to
84 age group. The age-ratios were .97 for the 1-to—9-cigarettes
per day group, 1.19 for the 10-to-19 per day group, 1.45 for the
20-39 group, and 1.63 for the 40-plus group. For all cigarette
Smokers the age-adjusted mortality ratio was 1.26. By age
group, mortality ratios were 1.12 for the 35-to—44 age group,

55
TABLE 6.—Mortality ratios for female cigarette smokers by
number of cigarettes smoked per day and age;

 

 

 

 

British females

Number of
cigarettes Age Total, 45-74
per day 45-54 55-64 65-74 Age-adjusted!
Nonsmokers 1.00 1.00 1.00 1.00
<20 1.49 1.09 79 1.08

20+ 1.85 1,51 1.55 1.60
All Smokers 1.66 1.25 .98 1.25

 

1Adjusted by the direct method using as standard the age distribution of all

women.
SOURCE: National Heart, Lung, and Blood Institute (10).

TABLE 7.—Mortality ratios for female cigarette smokers by
number of cigarettes smoked per day and age;
Norwegian females

 

 

 

Number of
cigarettes Age Total, 45-74
per day 45-64 65-74 Age-adjusted?
Nonsmokers 1.00 1.00 1.00
<20 1.54 1.07 1.33

20+ 1,41 89 1.18
All smokers 1.49 1.02 1.28

 

‘Adjusted by the direct method using as standard the age distribution of all
women.
SOURCE: National Heart, Lung, and Blood Institute (10).

1.31 for the 45-to-54 age group, 1.27 for the 55-to-65 group,
1.31 for the 65-to~74 group and 1.14 for the 75-to—84 age group.

Data from the Swedish study (Table 3) appear to be rea-
sonably consistent with the ACS data in Table 2. The 1-to-7-
cigarettes—per-—day group had an age-adjusted mortality ratio
of 1.0 (compared with .97 for the 1-to~9 group above) and 2.0 for
the 16-plus group (compared with 1.63 for the 40-plus group
above). For three of the four age groups, the mortality ratios
were directly associated with level of smoking. By age group,
the highest mortality ratios were observed for the two youngest
age groups and the lowest for the two oldest groups. The overall]
ratio for all cigarette smokers was 1.2.

For the other studies (Tables 4-7) mortality patterns were
generally similar in that mortality ratios tended to be highest

56
TABLE 8.—Mortality ratios for female cigarette smokers by
number of cigarettes smoked per day; females in the
British Doctors Study

 

 

 

Number of

cigarettes Total,

per day Age-adjusted?

Nonsmokers 1.00
1-14 0.94

15-24 1.54

25+ 1.66

All Smokers 1.23

 

1Based on annual death rates standardized for age.
SOURCE: Cederlof, R. (2).

with heaviest smoking and tended to be lowest at the oldest
ages.

For the Japanese study and the British Doctors Study, mor-
tality ratios by amount smoked and age were not reported.
However, an overall age-adjusted mortality ratio for female
cigarette smokers was reported in the Japanese study, while in
the British Doctors Study this ratio was obtained from unpub-
lished data based on 22 years of follow-up (Table 8). We list these
along with the overall ratios for the other studies:

Total mortality ratio

 

Study age-adjusted
American Cancer Society 1.26
Swedish 1.20
Canadian 1.31
Japanese 1.28
British Doctors 1.23
Framingham 1.43
British Migrants 1.25
Norwegian Migrants 1.28

All ratios here are greater than unity. The largest ratio is 1.43
for Framingham. The other seven ratios are close to one
another, ranging from 1.2 for the Swedish study to 1.31 for the
Canadian study.

DURATION OF SMOKING

Overall mortality ratios for women increased with duration of
the smoking habit based on data from the Canadian and

57
TABLE 9.— Age-adjusted mortality ratios of female cigarette
smokers, by number of cigarettes smoked per day
and age began smoking; subjects aged 45-54 at start
of study. 25-State Study

 

 

 

Number of Age began smoking
cigarettes
per day 25+ 15-24
Nonsmokers 1.00 1.00
1-9 0.95 0.88
10-19 : 1.17 1.23
20-39 1.33 1.61
40+ ** 1.85

 

**Ratio not shown—less than 10 expected deaths.
SOURCE: Hammond, E.C. (5).

TABLE 10.— Age-adjusted mortality ratios of female cigarette
smokers, by number of cigarettes smoked per day
and degree of inhalation. Subjects aged 45~—54 at
start of study. 25-State Study

 

 

 

Number of Degree of inhalation of smoke

cigarettes

per day None— Slight Moderate— Deep
1-9 0.85 1.04

10-19 1.27 1.17

20~39 1.41 1.58

40+ ** 2.19

 

**Ratio not shown—less than 10 expected deaths.
SOURCE: Hammond, E.C. (5).

Swedish studies (1,4). Among Canadian women who smoked for
10 or more years the mortality ratio, adjusted for age, was 1.37
compared to a ratio of 1.08 for women smoking less than 10
years. In the Swedish study an excess risk was found for women
smoking 30 or more years (1.4). For those smoking less than 30
years the ratio was 1.0.

AGE BEGAN SMOKING

Table 9 shows mortality ratios for women who were 45 to 54 by
number of cigarettes smoked per day and age began smoking
(5). Except for the light cigarette smokers (1-to-9-per-day),
those taking up the habit at ages 15 to 24 had higher mortality
ratios than those who started smoking at older ages.

58
TABLE 11.— Age-adjusted mortality ratios of female cigarette
smokers, by number of cigarettes smoked per day
and degree of inhalation and age. 25-State Study

 

 

 

Degree Age

of

Inhalation 35-44 45-54 55-64 65-74 75-84
Nonsmokers 1.00 1.00 1.00 1.00 1.00
None * 1.01 1.11 1.12 0.96
Slight 1.22 1.21 1.28 1.26 1.21
Moderate 1.05 1.30 1.32 1.41 *
Deep 1.40 1.78 1.64 “* +*

 

**Ratio not shown—less than 10 expected deaths.
SOURCE: Hammond, E.C. (5).

Mortality data for women smokers, according to age started,
are also available from the Swedish study (1); age-adjusted
ratios were reported as 1.7, 1.6, and 1.1 for age started less than
17, 17 to 18, and 19 plus, respectively.

INHALATION

Table 10 shows mortality ratios for female cigarette smokers
who were 45 to 54 years of age according to number of cigarettes
smoked per day and degree of inhalation of smoke (5). No clear
pattern emerges. The “moderate-deep” group had higher mor-
tality ratios than the “none-slight” group in two of three com-
parisons.

Table 11 shows mortality ratios for female cigarette smokers
by degree of inhalation and age (5). A fairly consistent general
pattern emerges; mortality ratios vary directly with degree of
inhalation. This is seen in each age group, except perhaps the
35-to-44 age group.

Mortality data for female cigarette smokers according to in-
halation are also available from the Swedish study (1); age-
adjusted ratios were reported as 1.1, 1.2, and 1.6 for the no inha-
lation, light inhalation, and deep inhalation groups, respec-
tively.

“TAR” AND NICOTINE CONTENT OF CIGARETTES

The relationship between overall mortality and the “tar” and
nicotine content of cigarette smoke was recently examined by
Hammond, et al. (6). In this study, “tar” and nicotine levels
(T/N) were defined as follows: “high” T/N, 25.8 to 35.7 mg “tar”
and 2.0 to 2.7 mg nicotine; “medium” T/N, 17.6 to 25.7 mg “tar”

59
TABLE 12.— Adjusted mortality ratios for males and females, by
“tar” and nicotine content of cigarettes usually

 

 

 

 

smoked
Mortality Ratios
“High” “Medium” “Low”
Sex TIN TIN TIN
Males 1.00 0.94 0.85
Females 1.00 0.88 0.83
Total 1.00 0.91 0.84

 

SOURCE: Hammond, E.C. (6).

TABLE 13.— Adjusted mortality ratios for males and females
smoking low “tar” and nicotine cigarettes and
subjects who never smoked regularly

Mortality ratios

 

 

Sex “Low” T/N Nonsmokers
Males 1.00 0.61
Females 1.00 0.74
Total 1.00 0.66

 

SOURCE: Hammond, E.C. (6).

TABLE 14.— Overall mortality ratios of cigarette smokers
compared to nonsmokers, by sex and by “tar” and
nicotine content of cigarettes usually smoked

 

 

 

Non- “Low” “Medium” “High”
Sex smokers TIN TIN TIN
Males 1.00 1.66 1.85 1.96
Females 1.00 1.37 1.45 1.65
Total 1.00 1.52 1.64 1.80

 

SOURCE: Hammond, E.C. (6).

and 1.2 to 1.9 mg nicotine; “low” T/N, less than 17.6 mg “tar” and
less than 1.2 mg nicotine.

Table 12 shows the overall mortality ratios of male and female
smokers by these “tar” and nicotine levels. In this instance, the
mortality ratio of the “high” T/N smokers was represented as
1.00 to illustrate the reduction in overall mortality that occurred
with lower T/N cigarettes. There was a small reduction in the
risk of dying with the use of lower T/N cigarettes. The mortality

60
ratio was reduced to 0.91 for the ‘‘medium” T/N smokers and
was further reduced to 0.84 for the “low” T/N smokers. The mor-
tality ratios were lower for women than for men.

In a separate analysis, a comparison was also made between
the mortality ratios of “low” T/N smokers and nonsmokers.
These data are presented in Table 13. The mortality ratio of the
“low” T/N group was designated as 1.00. Nonsmokers had over-
all mortality ratios that were considerably less than those of
“low” T/N smokers.

The combined data from Tables 12 and 13 are shown in Table
14 where mortality ratios were calculated using nonsmokers as
the reference. Combining these data from two separate
analyses that are not exactly comparable results in figures that
are only approximate.

Hammond also compared death rates of smokers of relatively
few (1 to 9) “high” T/N cigarettes with those of smokers who
smoked relatively large numbers (20 to 39) of “low” T/N cigar-
ettes (17). The death rates of these two groups were very simi-
lar.

Comments

Mortality ratios for women who smoke cigarettes ranged from
1.2 in the Swedish study to 1.43 in the Framingham study. As
with men, mortality ratios for women who smoke cigarettes var-
ied directly with amount smoked, depth of inhalation, “tar” and
nicotine content of the cigarette and duration of smoking, and
varied inversely with the age when smoking was started.

In attempting to study cigarette smoking and mortality
among women, a major difficulty is the lack of large-scale
epidemiological studies addressed specifically to female popula-
tions. The main findings of this review depend heavily on one
study, that of the American Cancer Society. For the other
studies reviewed here, the numbers of women—and of deaths
among them—are often too sparse to permit meaningful statis-
tical analyses. Thus, for example, little can be said about the
survivorship experience of women who give up cigarette smok-
ing. We strongly recommend, where possible, extending the
length of follow-up of women who are already enrolled in these
prospective studies. It is also highly recommended that new
studies be conducted that are specifically addressed to women
and smoking-related mortality.

Summary
1. The mortality ratio for women who smoke cigarettes is
about 1.2 or 1.3.
61
2. Mortality ratios for women increase with the amount
smoked. In the largest prospective study the mortality ratio
was 1.68 for the two-pack-a-day smoker as compared to
nonsmokers.

3. Mortality ratios are generally proportional to the duration
of cigarette smoking; the longer a woman smokes, the greater
the excess risk of dying.

4. Mortality ratios tend to be higher for those women who
begin smoking at a young age as compared to those who begin
smoking later.

5. Mortality ratios are higher for those women who report
they inhale smoke than for those who do not inhale.

6. Mortality ratios for women tend to increase with the tar
and nicotine content of the cigarette.

7. Mortality ratios for female smokers are somewhat less
than for male smokers. This may reflect differences in exposure
to cigarette smoke, such as starting smoking later, smoking
cigarettes with lower “tar” and nicotine content, and smoking
fewer cigarettes per day than men.

8. Women demonstrate the same dose-response relationships
with cigarette smoking as men. An increase in mortality occurs
with an increase in number of cigarettes smoked per day, an
earlier age of beginning cigarette smoking, a longer duration
of smoking, inhalation of cigarette smoke, and a higher “tar”
and nicotine content of the cigarette. Women who have smok-
ing characteristics similar to men may experience mortality
rates similar to men.

References

(1) BEST, E.W.R. A Canadian Study of Smoking and Health. Department of
National Health and Welfare, Epidemiology Division, Health Services
Branch, Biostatistics Division, Research and Statistics Directorate,
1966, 137 pp.

(2) CEDERLOF, R., FRIBERG, L., HRUBEC, Z., LORICH, U. The Rela-
tionship of Smoking and Some Social Covariables to Mortality and
Cancer Morbidity. A Ten Year Follow-up in a Probability Sample of
55,000 Swedish Subjects Age 18 to 69. Part I and II. Stockholm,
Karolinska Institute, Department of Environmental Hygiene, 1975,
201 pp.

(3) DOLL, R., GRAY, R., PETO, R. Mortality in Relation to Smoking: Ob-
servations on Female Doctors. (Unpublished manuscript)

(4) DORN, H.F. The relationship of cancer of the lung and the use of to-
bacco. The American Statistician 8(5): 7-13, December 1954.

(5) HAMMOND, E.C. Smoking in relation to the death rates of one million
men and women. In: Haenszel, W. (Editor). Epidemiological Ap-
proaches to the Study of Cancer and Other Chronic Diseases, National
Cancer Institute Monograph 19. Department of Health, Education,
and Welfare, Public Health Service, National Cancer Institute,
January 1966, pp. 127-204.

62
(6)

(7)

(8)

(9)

(10)

dy

(12)

(13)

(14)

(15)

(16)

(17)

HAMMOND, E.C., GARFINKEL, L., SEIDMAN, H., LEW, E.A. “Tar”
and nicotine content of cigarette smoke in relation to death rates. En-
vironmental Research 12(3): 263-274, December 1976.

HIRAYAMA, T. Smoking in relation to the death rates of 265,118 men
and women in Japan. A report of 5 years of follow-up. Presented at the
American Cancer Society's 14th Science Writers’ Seminar, Clearwater
Beach, Florida, March 24-29, 1972, 15 pp.

NATIONAL CENTER FOR HEALTH STATISTICS. Mortality from
Diseases Associated with Smoking: United States, 1950-1964. De-
partment of Health, Education, and Welfare, Public Health Service,
National Center for Health Statistics, Public Health Service Publica-
tion No. 1000-Series 20, 50. 4, October 1966, 45 pp.

NATIONAL CENTER FOR HEALTH STATISTICS, Office of Health
Research, Statistics, and Technology, Public Health Service, Depart-
ment of Health Service, Department of Health, Education, and Wel-
fare. (Unpublished data)

NATIONAL HEART, LUNG, AND BLOOD INSTITUTE. Proceedings
of the Conference on the Decline in Coronary Heart Disease Mortality.
Department of Health, Education, and Welfare, Public Health Service,
National Institutes of Health, NIH Publication No. 79-1610, May 1979,
399 pp.

NATIONAL HEART, LUNG, AND BLOOD INSTITUTE. Some charac-
teristics related to the incidence of cardiovascular disease and death:
Framingham Study, 18-year follow-up. In: The Framingham Study: An
Epidemiological Investigation of Cardiovascular Disease. Kannel,
W.B., Gordon, T. (Editors). DHEW Publication No. (NIH) 74-599, Feb-
ruary 1974.

NATIONAL HEART, LUNG, AND BLOOD INSTITUTE. Department
of Health, Education, and Welfare, Public Health Service, National
Institutes of Health, 1979. (Unpublished data)

PEARL, R.B., LEVINE, D.B., GERSON, E.J. Studies of Disease Among
Migrants and Native Populations in Great Britain, Norway, and the
United States. II. Conduct of Field Work in the United States. National
Cancer Institute Monograph 19. Department of Health, Education,
and Welfare, U.S. Public Health Service, National Cancer Institute,
1966, pp. 301-320.

REID, D.D. Studies of Diseases Among Migrants and Native Popula-
tions in Great Britain, Norway, and the United States. I. Background
and Design. National Cancer Institute Monograph 19. Department of
Health, Education, and Welfare, Public Health Service, National
Cancer Institute, 1966, pp. 287-199.

REID, D.D., CONFIELD, J.. MARKUSH, R.E., et al. Studies of Disease
among Migrants and Native Population in Great Britain, Norway, and
the United States. III. Prevalence of Cardiorespiratory Symptoms
Among Migrants and Native Born in United States. National Cancer
Institute Monograph 19. Department of Health, Education, and Wel-
fare, Public Health Service, National Cancer Institute, 1966, pp. 321-
346.

ROGOT, E. Cardiorespiratory disease mortality among British Norwe-
gian migrants to the United States. American Journal of Epidemiology
108(3): 181~191, 1978.

U.S. PUBLIC HEALTH SERVICE. Smoking and Health. A Report of
the Surgeon General. Department of Health, Education, and Welfare,
Public Health Service, Office of the Assistant Secretary for Health,
Office on Smoking and Health. DHEW Publication No. (PHS) 79- 50066,
1979, 1251 pp.

63
MORBIDITY.
MORBIDITY

The relationship between cigarette smoking and morbidity
has been summarized in the 1979 Surgeon General’s Report.
That report contained data from the National Center for Health
Statistics Health Interview Survey (HIS) showing the relation-
ship for both men and women between smoking and the preva-
lence of selected chronic diseases, the incidence of acute illness,
days lost from work, days of bed disability, and perceived health
status. This section will present additional data from the Health
Interview Survey on trends in days lost from work and limita-
tion of activity.

Days Lost from Work

Workers who smoke report losing more work days due to ill-
ness and injury than do nonsmokers. This relationship has been
observed for both men and women every year that the National
Health Interview Survey has included questions on cigarette
smoking. For example, in 1965 working women who smoked re-
ported 6.6 work-loss days; working women who had never
smoked reported only 4.8 work-loss days (see Table 1). Similarly,
in the 1977 HIS women who smoked reported 6.6 days lost from
work compared to 5.7 days lost from work by those who never
smoked.

The National Clearinghouse for Smoking and Health used the
earlier 1965 data to estimate the number of “excess” days lost
from work among cigarette smokers. This estimation was ob-
tained by calculating the expected number of work-loss days if
all workers had the same work-loss experience as those who had
never smoked cigarettes. It was estimated that approximately
20 percent of all work-loss days due to illness and injury could be
attributed to the higher rates of loss among current and former
smokers (2). The 1979 Surgeon General’s Report presented simi-
lar calculations, based on 1974 data, and again the estimate was
about 20 percent of all work-loss days. These calculations were
not sex specific. Certain modifications in the collection proce-
dures have lowered the male response rate for the smoking data
and may, thus, make comparisons of more recent data by sex
less than ideal. However, the data do show that in 1977 the
work-loss rate among women who never smoked was higher
than in 1965, while the rates among current smokers remained
about the same. This would tend to reduce the number of “‘ex-
cess” days among women attributable to smoking. There has
been a slight decrease in work loss among males who never
smoked. Former smokers reported fewer work-loss days in 1977
than in 1965. Although the difference in work-loss days between

67
TABLE 1.—Days lost from work per year due to illness and
injury, per currently employed persons 17 years
old and older, by smoking status, sex and age:
United States, 1965 and 1977

 

Present Former Never
Total! Smoker Smoker Smoked

 

Percent -of work-loss days

1965
Female
17+4 5.6 6.6 6.7 4.8
17-44 5.5 6.6 6.0 4.5
45-64 6.0 6.7 7.4 5.3
Male
17 +8 5.7 5.9 6.8 4.6
17-44 4.1 4.7 3.6 3.4
45-64 7.8 79 9.8 5.6
1977
Female
20 +3 6.0 6.6 5.4 5.7
20-44 6.1 6.8 5.4 5.4
45-64 6.4 6.5 5.9? 6.5
Male
20 +3 5.3 5.9 6.1 4.2
20-44 5.1 6.0 5.5 4.4
45-64 5.6 5.9 6.2 3.9

 

‘Includes unknown smoking status.

?Figure does not meet standards of reliability or precision.
5Includes ages 65 and over.

SOURCE: National Center for Health Statistics (1).

1965 and 1977 is small, it could be attributed to the assumption
that in recent years the former smoker groups have a greater
proportion of people who stopped smoking for preventive rea-
sons, that is, before they had experienced serious health conse-
quences.

Further study is needed to determine the association between
“excess” days lost from work by smokers and specific diseases.
Such an analysis would help explain the economic impact of
smoking in the work place.

Limitation of Activity

The Health Interview Survey also regularly collects data on
the long-term impact of chronic illness. Respondents were asked
if chronic illness limited their activities (3). Estimates of the
percent of the population with limitation of activity by cigarette

68
smoking status are shown in Table 2 for 1965 and 1977. Detailed
interpretation of trend data is difficult; however, there appears
to be a relationship between smoking and the impact of chronic
illness. In general, the 1977 data indicate that women under 65
who have ever smoked are more likely to have a limitation of
activity than those who never smoked. There are no marked
differences between current and former smokers. Among eld-
erly women in 1977, there were no differences in limitations of
activity by smoking status.

TABLE 2.— Percent of persons with limitation of activity due to
chronic conditions, by cigarette smoking status, sex
and age: United States, 1965 and 1977

 

Present Former Never
Total! Smoker Smoker Smoked

 

Percent with limitation

1965
Female
17+ 17.3 12.7 17.3 19.8
17—44 8.3 8.8 9.8 7.7
45-64 19.5 17.4 22.1 20.2
65+ 45.1 39.8 48.6 45.4
Male :
17+ 17.3 15.3 23.0 17.7
17-44 7.3 7 8.0 6.2
45-64 20.0 20.9 22.1 15.7
65 + 53.7 52.7 56.3 52.9
1977
Female
20 + 17.6 16.0 18.1 18.3
20-44 8.0 9.2 8.4 7.0
45-64 21.5 24.2 23.9 19.8
65 + 89.2 36.3 35.5 38.8
Male
20 + 20.0 20.5 24.1 17.6
20-44 9.6 12.4 8.3 75
45-64 25.7 27.5 25.7 25.7
65 + 47.5 52.7 47.6 42.5

 

‘Includes known smoking status.
SOURCE: National Center for Health Statistics (1).

Cigarette Smoking and Occupation*

The Health Interview Survey provides a considerable data
base on cigarette smoking behavior and occupational status.

*See: “Interaction Between Smoking and Occupational Exposures” in this
Report.

69
The data are available from a national probability sample of
about 40,000 households for the years 1965, 1966, 1970, 1974,
1976, 1977, 1978, and 1979. However, only minimal analysis has
been conducted on this potentially valuable data base (4). This
brief section presents data on smoking patterns for only two of
these periods—1970 and 1976. Researchers are encouraged to
investigate these data more fully through the purchase of pub-
lic use data tapes (1). The importance of this data base increases
as new evidence becomes available on the increased health risks
experienced by smokers in certain occupations. The problems of
relatively small sample sizes in high-risk occupations can be
partially overcome by combining several years of the HIS data
tapes.

Tables 3 and 4 show smoking characteristics of broad occupa-
tional groups—i.e., white collar, blue collar, service and farm
workers —for 1970 and 1976, respectively. Service and blue col-
lar workers, both women and men, are more likely to smoke
than are white collar and farm workers, but the differences are
much less among female workers. In 1970, there were virtually
no differences among female white collar, blue collar, and serv-
ice workers; more recently, however, there has been a slight
increase in smoking among the latter two groups. Caution
should be used in drawing conclusions from these data based on
differences of only a few percentage points since such dif-
ferences can be well within sampling error. White collar work-
ers who smoke tend to be heavier smokers than other types of
workers, and this pattern is more marked among female white
collar workers.

The proportions of cigarette smokers by more detailed occu-
pational classes are shown in Tables 5 and 6 for 1970 and 1976.
Within three of four subgroups of white collar workers —
professionals, managers, and sales people—the proportion of
smokers among women is the same as for men in the same occu-
pational group. This also appears to be true for laborers, who
show the highest levels of smoking among both women and men.

Summary

The 1979 Report of the Surgeon General summarized the in-
formation on smoking and morbidity as follows:

1. In general, female current cigarette smokers report more
acute and chronic conditions including chronic bronchitis
and/or emphysema, chronic sinusitis, peptic ulcer disease, and
arteriosclerotic heart disease, than women who never smoked.

2. There is a dose-response relationship between the number

70
TL

TABLE 3.—Percent distribution of the population 17 years and over by cigarette smoking status, according to
sex and occupation category, United States, 1970

 

 

 

 

 

 

Sex and
occupation category Percent distribution
Total Never Former Present Present smokers— no. of cigarettes per day?
population! smoked smokers smokers Total? <15 15-24 25+
Female
Total population 100.0 54.0 11.2 34.9 100.0 39.3 42.4 18.2
Total currently employed 100.0 54.3 111 34.6 100.0 38.7 43.3 18.0
White collar workers 100.0 53.2 12.6 34.2 100.0 37.6 42.8 19.6
Blue collar workers 100.0 55.1 8.5 36.5 100.0 40.7 44.4 14.9
Service workers 100.0 55.7 9.2 35.2 100.0 41.6 41.0 17.4
Farm workers 100.0 74.3 *7.5 18.6 100.0 *49.2 *33.3 *19.0
Male

Total population 100.0 28.8 24.9 46.2 100.0 25.8 45.1 29.1
Total currently employed 100.0 28.8 25.2 46.0 100.0 25.5 45.3 29.3
White collar workers 100.0 31.6 29.1 39.3 100.0 23.8 43.4 32.8
Blue collar workers 100.0 24.8 22.4 §2.8 100.0 25.5 46.4 28.0
Service workers 100.0 31.1 20.8 48.1 100.0 31.1 43.3 25.6
Farm workers 100.0 40.7 24.8 34.4 100.0 35.5 45.1 19.4

 

‘Excludes unknown if ever smoked.

?Excludes unknown amount of cigarettes smoked.

*Figure does not meet standards of reliability or precision.
SOURCE: National Center for Health Statistics (1).
oL

TABLE 4— Percent distribution of the population 20 years and over by cigarette smoking status, according to
sex and occupation category, United States, 1976

 

 

Sex and Total Never Former Present Present smokers—no. of cigarettes per day?
occupation category population! smoked smokers smokers Total? <15 15-24 25 +
Female

Total population 100.0 54.3 13.8 32.0 100.0 36.5 43.8 19.6

Total currently employed 100.0 50.8 13.3 35.9 100.0 36.5 44.0 19.5
White collar workers 100.0 61.1 14.6 34.3 100.0 35.3 42.4 22.3
Blue collar workers 100.0 50.7 10.2 39.0 100.0 38.0 44.3 17.6
Service workers 100.0 49.1 11.9 39.0 100.0 37.9 48.3 13.7
Farm workers 100.0 59.8 _ * 31.3 100.0 34.6 * *

Male

Total population 100.0 29.2 28.9 41.9 100.0 24.2 44.8 31.1

Total currently employed 100.0 29.5 27.1 43.4 100.0 21.9 45.4 32.8
White collar workers 100.0 34.0 29.4 36.6 100.0 20.8 43.6 35.6
Blue collar workers 100.0 24.3 25.3 50.4 100.0 21.2 47.4 31.5
Service workers 100.0 29.4 23.4 47.2 100.0 27.6 40.0 32.4
Farm workers 100.0 34.9 28.2 36.9 100.0 29.4 44.9 25.7

 

‘Excludes unknown if ever smoked.

*Excludes unknown amount of cigarettes smoked.

*Figure does not meet standards of reliability or precision.
SOURCE: National Center for Health Statistics (1).
TABLE 5.— Estimates of the percentage of current, regular cigarette smokers, adult ages 17 years and over,
according to labor force status, occupation, and sex, United States, 1970

 

 

Female Male
Total Total
17+ 17-44 45-64 17+ 17-44 45-64
Total 34.9 36.8 33.7 46.2 49.0 44,4
Currently employed 34.6 36.4 33.7 46.0 48.7 44,1
White collar total 34.2 34.9 34.3 39.3 41.1 38.4
Professional, technical
and kindred 28.1 29.4 26.3 31.7 32.8 30.6
Managers & administrators
except farm 40.8 48.4 38.3 42.8 47.4 40.0
Sales workers 34.6 35.3 35.7 44,9 46.8 46.1
Clerical & kindred workers 35.8 35.9 36.4 43.3 45.2 41.5
Blue collar total 36.5 39.9 33.5 52.8 56.1 49.2
Craftsmen & kindred
workers 40.4 44.4 37.0 51.7 56.1 47.2
Operatives and kindred
workers 36.5 40.0 33.5 54.7 57.5 50.7
Laborers, except farm *23.3 *25.6 *20.9 50.9 52.0 52.9
Service 35.2 39.3 33.5 48.1 48.3 51.7
Farm 18.6 *25.9 *15.5 34.4 38.7 37.7
Unemployed 38.4 40.8 32.9 52.3 54.4 53.0
Homemakers 29.7 37.3 32.3 NA NA NA

 

NOTE: Unknown if ever smoked excluded from calculation.
4 *Figure does not meet standards of reliability or precision.
oo SOURCE: National Center for Health Statistics (1).
= TABLE 6.— Estimates of the percentage of current, regular cigarette smokers, adults ages 20 years and over,

according to labor force status, occupation, and sex, United States, 1976

 

 

Female Male
Total Total
20+ 20-44 45-64 20+ 20-44 45-64
Total 32.0 36.9 34.8 41.9 47.6 41.3
Currently employed 35.9 37.0 36.1 43.4 46.8 39.7
White collar total 34.3 33.8 36.9 36.6 38.6 35.3
Professional, technical
and kindred 29.1 28.6 32.7 30.0 31.1 29.9
Managers & administrators
except farm 41.6 42.7 40.8 41.0 46.4 36.1
Sales workers 38.1 37.0 42.6 39.9 42.6 38.0
Clerical & kindred workers 34.8 34.7 36.0 40.4 40.1 44.2
Blue collar total 39.0 43.7 33.6 50.4 54.1 44.3
Craftsmen & kindred workers 40.5 46.9 35.6 48.0 52.1 41.6
Operatives and kindred
workers 37.6 42.5 31.2 52.3 55.3 46.2
Laborers, except farm 56.3 52.6 * 53.7 56.9 51.7
Service 39.0 42.8 37.2 47.2 51.1 44.8
Farm 31.3 51.0 * 36.9 45.4 35.0
Unemployed 40.0 41.0 39.2 56.8 59.9 53.8
Usual activity —homemakers 29.0 37.1 32.2 NA NA NA

 

NOTE: Unknown if ever smoked excluded from calculation.
“Figure does not meet standards of reliability or precision.
SOURCE: National Center for Health Statistics CD.
of cigarettes smoked per day and the frequency of reporting for
most of the chronic conditions.

3. The age-adjusted incidence of acute conditions (e.g., in-
fluenza) for women smokers is 20 percent higher for women who
had ever smoked than for nonsmokers.

Additional data from the Health Interview Survey (HIS) is
presented:

1. Currently employed women who smoke cigarettes report
more days lost from work due to illness and injury than working
women who do not smoke.

2. Limitation of activity is reported more commonly among
women under the age of 65 who have ever smoked than among
those who never smoked.

References

(1) NATIONAL CENTER FOR HEALTH STATISTICS. Standardized
Micro-Data Tape Transcript. Department of Health, Education, and
Welfare, Public Health Service, DHEW Publication No. 781-213, June
1978.

(2) NATIONAL CLEARINGHOUSE FOR SMOKING AND HEALTH. Smok-
ing and Illness. Department of Health, Education, and Welfare, Public
Health Service, Bureau of Disease Prevention and Environmental Con-
trol, National Center for Chronic Disease Control, National Clearing-
house for Smoking and Health, PHS Publication No. 1662, July 1967, 6

pp.

(3) WILDER, C.S. Limitation of activity due to chronic conditions, U.S. 1974.
Department of Health, Education, and Welfare, Public Health Service,
Health Resource Administration, National Center for Health Statistics,
Series 10, No. 111, Public Health Service Pub. No. (HRA) 77-1537, June
1977, 65 pp.

(4) WILSON, R.W. Cigarette smoking, disability days and respiratory condi-
tions. Journal of Occupational Medicine 15(3): 236-240, March 1973.

75