controlled trials offer a solution to this problem, because they test a
smoking cessation intervention in the most rigorous way possible
using human subjects (2, 37, 70). However, even in such trials, those
assigned to the no-intervention group modify their smoking habits,
and those assigned to the intervention group have incomplete
success in quitting.

Ideally, each risk factor should be treated independently, but
modification of one risk factor oftén results in changes in other risk
factors. For instance, some studies have noted that cigarette smoking
cessation can lead to a modest weight gain (3, 26, 47). In a good
primary prevention program there would be an effort to reduce or to
eliminate the weight gain that sometimes accompanies cessation.
Although the multifactorial approach is less precise, it has been
considered to be a more practical approach to the problem. There has
been only one trial of smoking cessation per se (50, 51). —

Two types of primary prevention trials involving smoking cessa-
tion are underway. The first type of study, exemplified by the
Multiple Risk Factor Intervention Trial (MRFIT), selected subjects
at high risk of CHD based on a combination of cigarette smoking
history, elevated blood pressure, and high serum cholesterol level
(43). Subjects were then randomized into either a special treatment
group or a comparison group. The Whitehall study (50, 51) and the.
Oslo study (27) are also of this type.

In the second type of study, communities or groups rather than
individuals are randomized into a treatment or a control group. The
WHO heart disease study randomized men according to the factory
where they work and followed the individual factory workers (70).
The North Karelia study randomized two separate communities in
Finland (48). Whereas the factory workers were individually fol-
lowed, the communities were monitored only in a cross-sectional
manner and individuals were not followed longitudinally.

The primary hypothesis in these studies is that reduction of the
risk factors will reduce the incidence of and mortality from CHD.
The first step in testing this hypothesis requires that the subjects or
groups be successfully recruited and categorized at entry to the study
and that a very high percentage be successfully followed for the
duration of the study.

The second step in testing the hypothesis requires the successful
reduction of the major risk factors—smoking, high blood pressure,
and high serum cholesterol. Often a selected subsample of high risk
subjects receives more intensive individual intervention, but the rest
of the treatment group receives only community health education. It
is not known how large a reduction in risk factors is necessary to
observe a decrease in CHD, except that the larger the reduction in
risk factor the greater the chance for a decrease in CHD. Specific
goals for reduction of risk factors can be based on the presumption

301
that such reduction would result in a statistically significant
decrease in the incidence of and mortality from CHD according to
the observational epidemiologic studies. The relationship between
the reduction of this risk score and actual reduction in the frequency
of disease is, of course, the hypothesis being tested.

The third and most critical step depends on the first two: that is,
the measurement of outcome—changes in the incidence of or death
from CHD. The ability to measure the incidence requires careful and
unbiased monitoring of the sample. The determination of total and
cardiovascular mortality is much simpler, since it depends only on
minimizing the number of participants lost to followup. The commu-
nity studies attempt to compare the death rates between two or more
communities, and the power of such a statistical test is obviously
very weak.

Both the community studies and the individualized studies are
also confounded by the uncertainty of the interval between risk
factor change and reduction in risk of disease. Those subjects most
likely to die or to have a heart attack in the first few years of the
study are those with the most extensive disease at baseline. Unless
the population is followed long enough to include both the lag period
and the effects of the initial selection of those with advanced
subclinical disease, a spurious interpretation of the study results is
possible.

Generally, studies of experimental communitywide interventions
are unlikely to determine the efficacy of smoking cessation on
reducing the incidence of and mortality from CHD because of the
difficulty in determining the effects of smoking cessation on specific
individuals in the community and in separating out the effect of
smoking cessation froni other changes in the community. In many of
these studies the percentage of men who reported quitting smoking
is relatively small, which reduces the power of the study at least in
terms of smoking cessation.

This review focuses on all intervention studies involving smoking
cessation for which CHD mortality outcome data have been pub-
lished. In another section of this Report a detailed review on
smoking cessation in clinical and community trials is presented.
Discussed in detail here are data from the U.S. Multiple Risk Factor
Intervention Trial (43), the Whitehall study in London, England (51),
the Oslo study in Norway (27), the WHO European Collaborative
study (69), and the North Karelia project in Finland (48). Omitted
from this section are smoking intervention studies without published
mortality outcome data, such as the Stanford, California, three-
community study (16), the Stanford five-community study (25), the
Goteborg, Sweden, study (64, 65, 66), and the Helsinki, Finland,
study (46).

302
Essential details from each trial are described in the following
pages and tables. The initial population characteristics and risk
factor changes are summarized in Table 4. The mortality outcomes
in the intervention and control groups are summarized in Table 5 for
coronary heart disease and total mortality. Available results are also
given for other circulatory diseases, lung cancer, and other cancer.
The results for the four trials involving cohort mortality followup
are combined in Table 6. Coronary heart disease incidence rates
from the Oslo and WHO studies are presented in Table 7. A
comparison of death rates in the MRFIT intervention and control
groups as a function of initial smoking status and status at 1-year
followup is summarized in Table 8. Observed deaths in the MRFIT
and Whitehall studies are compared with expected deaths based on
general population rates in Table 9. Comparisons of reductions in
CHD and total mortality from the observational studies and from the
MRFIT and Whitehall studies are made in Table 10.

Randomized Controlled Trials of Individuals
Multiple Risk Factor Intervention Trial

The Multiple Risk Factor Intervention Trial (MRFIT) was a
randomized controlled trial to investigate the effect of reducing
cardiovascular risk factors in a group of asymptomatic men at high
risk of cardiovascular disease (43). Out of 361,662 men initially
screened, 12,866 men aged 35 to 57 were selected for the trial
because they were at increased risk of death from CHD, but without
clinical evidence of CHD, and agreed to be randomized and reexa-
mined. A series of three complex screening procedures were used to
select the final 12,866 men, who constituted only 3.6 percent of those
screened. Men were designated as at increased risk because their
levels of three risk factors—cigarette smoking, serum cholesterol,
and blood pressure (BP)—were sufficiently high at a screening visit.
All of these men were in the upper 15 percent of a risk score
distribution based on data from the Framingham heart study; about
two-thirds were in the upper 10 percent of risk. For example, a man
whose diastolic BP was 90 mm Hg and who reported smoking 30
cigarettes per day was risk eligible at the 10 percent level if his
serum cholesterol level was at least 295 mg/dl. The study was
restricted to men, since including women, with their substantially
lower risk of CHD, would have necessitated a larger study popula-
tion.

The men were randomized into two groups of equal size and
identical baseline characteristics from December 1973 through
February 1976. A special intervention (SI) group of 6,428 men
received an intensive counseling program, aimed at cessation of
cigarette smoking, weight loss, and a change of diet for a reduction of
elevated serum cholesterol and BP levels. A usual care (UC) group of

302
TABLE 4.—Basic description of smoking cessation
intervention studies of males, including
demographic characteristics and risk factor

 

 

 

changes
Factory Community
Individual intervention intervention intervention
Variable MRFIT Whitehall Oslo WHO North Karelia
1972 1977
Number in intervention () group 6428 714 604 24,615 1834 «1785
Number in control (C) group 6438 731 628 25,169 2665 2616
Mean age (yrs) 46.2 52.9 45.2 48.5 ~42  ~A47
Age range (yrs) 35-57 40-59 40-49 40-49 25-59 30-64
Race (% white) 90 ~100 ~100 ~100 ~100
Followup period: Start date 12/1973 1968-70 1972-73 1971-76 1972
End date 2/1982 1979 1978 1982 1977
Average length (yrs) 7 10 5 6 5
Risk factor (RF) at start
Cigarette smokers (%) 59 100 79.4 60 51.5
Relative change '/end -29% -24% -12% 3.4%? -2.5%
Relative change/whole period -31% 54% -16% 1.9% ~?
Average no./day cigarettes 20 19.3 127 11.2 9.4
Relative change/end 30%? 33% 37% -10.1% -9.8%
Relative change/whole period -30%? 53% 45% 8.9% =
Serum cholesterol (mg/dl) 254 213° 329 217 263
Relative change/end 2% - -12% 0.5% 4.1%
Relative change/whole period -2% - -13% ~1.2% -
Blood pressure* (mm Hg) 91 -‘ -* 138 91.6
Relative change/end -4% _ _ -2,0% -2.8%
Relative change/whole period ~4% ~ _ -2.0% -
Combined change in CHD risk -22.2% - - -11.1% -17.4%

 

‘Relative change = (RF; - RFc)/RFc, except for North Karelia, where relative difference is determined from
(I-C) 1977 and (1-C) 1972.

* Estimated from available data.

* Not measured or not reported.

* Risk factor is not part of intervention.

* Diastolic, except for systolic in WHO study.

6,438 men received annual checkups including medical history,
physical examination, and laboratory studies at the MRFIT clinics,
but were referred to their personal physicians or other community
medical facilities for such treatment of their risk factors as was
considered individually appropriate. Thus, no intervention program
was offered to them. The results of their screening and annual
examinations were provided to their respective physicians who were
also informed as to the scientific objectives of the study.

The smoking intervention urged those SI participants who smoked
cigarettes to quit, but no effort was made to alter the smoking habits
of those who smoked only pipes and cigars (29). Dosage reduction—

304
Reported Thiocyanate-adjusted

cigarette smoking cigarette smoking

70 70

60 60
5 50 uc & 50 us
o o
r 40 x 40
uw w
a si a

30 30 si

oO 1 2 3 4 5 6 0 1 2 3 4 § 6
s
YEARS OF FOLLOWUP ‘ YEARS OF FOLLOWUP

FIGURE 1.—Mean risk factor levels for cigarette smoking
by year of followup for Multiple Risk Factor
Intervention Trial Research Group participants

NOTE: SI indicates special intervention
UC indicates usual care
S. indicates first screening visit
SOURCE: Multiple Risk Factor Intervention Trial Research Group (43).

switching to cigarettes low in tar and nicotine—was recommended
only as an intermediate step to cessation. Conventional behavior
modification techniques were used throughout the trial; aversive
techniques and hypnosis were used in selected instances. Ten-week
group sessions at the beginning of the trial and 5-day quit clinics
during the final years were found to be particularly successful
intervention approaches. Further details on the smoking interven-
tion are provided in an earlier section of this Report, and the serum
cholesterol and BP interventions are described elsewhere (43).

Statistically significant CHD risk factor reductions between the SI
and UC groups were obtained at each annual visit. Of particular
interest was the reduction in number of cigarette smokers, as shown
in Figure 1. At the beginning of followup, 59 percent of all men had
reported themselves as current cigarette smokers. At the 12-month
followup, the stated quit rates were 43 percent for SI men and 14
percent for UC men; at 72 months, the rates were 50 percent and 29
percent, respectively. Serum thiocyanate-adjusted quit rates at 12
months were 31 percent for SI men and 12 percent for UC men; at 72
months they were 46 percent and 29 percent, respectively. This
means that the SI group reduced its level of smokers 30 percent more
than the UC group. The risk factor changes are summarized in Table
4.

As of February 28, 1982, after an average period of followup of 7
years, there were 260 deaths among the UC men, of which 124 were
ascribed to CHD and 21 to other cardiovascular causes, as summa-

305
rized in Table 5. There were 265 deaths among the SI men, of which
115 were ascribed to CHD and 23 to other cardiovascular causes. The
key mortality endpoint of CHD was 7.1 percent less in the SI group
than in the UC group, while the death rate for all causes was 2.1
percent higher for the SI men. The approximate 90 percent
confidence interval (CI) for the percentage change in CHD mortality
attributable to MRFIT intervention ranges from a 25 percent
decrease to a 15 percent increase. There were 34 lung cancer deaths
in the SI group and 28 in the UC group and 47 other cancer deaths in
the SI group and 41 in the UC group.

The number of deaths in the UC group was substantially short of
expectation for the 6 complete years of followup as well as for the
average followup period of 7 years, as shown in Table 9. These
mortality patterns appear to be similar to those seen in healthy
persons selected for life insurance policies (22, 41). On the basis of
the design risk factor change assumptions and the Framingham risk
functions, 442 deaths (including 187 from CHD) were expected
among the 6,438 UC men by the end of 6 years of followup, but only
219 (including 104 from CHD) occurred (50 percent of expected);
about 515 deaths (including 220 from CHD) were expected after 7
years, but only 260 deaths (including 124 from CHD) occurred.

At least three possible explanations for these results must be
considered: (1) such an intervention program is without benefit in
terms of substantial decreases in mortality; (2) the intervention
program does affect CHD mortality, but the benefit was not observed
in this study, an effect of chance; or (3) one or more constituents in
the intervention program may have had an unfavorable effect on
survival in some subgroups, offsetting the beneficial effects of others.

Of these possible interpretations, a combination of favorable and
unfavorable effects of the intervention program seems most plausi-
ble to the MRFIT investigators. Even with the unexpected sizable
risk factor reduction among the UC men, the lower-than-expected
UC mortality, and the duration of intervention averaging only 7
years, the likelihood that these factors resulted in missing an overall
positive effect is relatively low. The data suggest that except for
some groups of hypertensive persons, particularly those with resting
ECG abnormalities, the MRFIT intervention is apparently associ-
ated with a lower CHD mortality in the SI group.

The MRFIT data also warrant analysis as an observational study
(Table 8). Of those who had been cigarette smokers at entry, 1,365
reported quitting at year 1 interview (and had confirmatory blood
SCN levels) and 6,298 did not quit. Over an average 6 years of
further followup, those who quit smoking at year 1 had a 1.10
percent CHD mortality rate, while those who continued smoking had
a CHD mortality rate of 2.03 percent. This corresponds to a relative
risk of 0.54 (1.10/2.03) or a 46 percent lower risk of CHD death for

306
TABLE 5.—Comparison of deaths (d) and death rates (r) in
the intervention and control groups of four
randomized controlled trials

 

 

 

Percentage
Intervention group Control group difference?®
Cause of death nt at Tc? deh (ri-re}/re
MRFIT: 7-year deaths
Coronary heart disease 0179 115 0193 124 -7.1
Other circulatory disease 0036 23 .0033 21 +98
Lung cancer 0053 34 0043 28 +216
Other cancer 0073 47 .0064 41 +148
All other causes 0072 46 0071 46 $0.1 |
All causes 0412 265 0404. = -260 +21
Whitehall: 10-year deaths
Coronary heart disease 0686 49 0848 62 -19.1
Other circulatory disease 0182 13 0164 12 +110
Lung cancer 0252 18 0328 24 ~23.2
Other cancer 0392 2B 0164 12 +139.0(p=.01)
All other causes 0210 15 0246 18 14.6
All causes 1723 123 ‘1761 128 -1.6
Oslo: 5-year deaths
Coronary heart disease
(Fatal MI and
sudden coronary death) 0099 6 0222 14 55.4
Other circulatory disease .0033 2 0016 1 +106.3
All cancer : .0083 5 0127 8 34.6
All other causes .0050 3 0016 1 +212.5
All causes 0265 16 0382 24 30.6
WHO: 6-year deaths
Coronary heart disease 0150 367 0162 355 1.4
All other causes . 0254 630 0253 569 +0.4
All causes 0404 997 0415 924 -2.7

 

' Death rate (r) equals deaths (d) divided by initial number in group.
* Unless indicated with a p value, differences are not statistically significant (p > 0.05), based on a two-tailed test
for a Poisson variable.

those who quit smoking compared with those who continued. For all-
cause mortality, the relative risk was 0.73 or a 27 percent lower risk
for those who quit compared with those who continued to smoke.

As demonstrated in subsequent portions of Table 8, when these
subjects are analyzed according to level of smoking at entry or by
status in the SI or the UC group, those who quit smoking always
enjoyed a substantially more favorable survival rate than those who
didn’t quit.

Thus, MRFIT data are entirely consistent with the numerous
previous studies showing that those who quit cigarette smoking
enjoy a substantially improved survival.

In conclusion, the MRFIT study has shown that it is possible to
apply an intensive long-term intervention program against three

307
coronary risk factors with considerable success in risk factor
changes. These results are accompanied by an apparent heterogenei-
ty of effects among sizable subgroups, and there must be caution in
reaching conclusions from such subgroup data. It may be relevant
that multifactor intervention received a less than optimal test,
owing in part to unexpected declines in risk factor levels and in part
to lower-than-expected mortality in the UC group. In regard to the
former, the UC men thus constituted to a considerable extent a
“treated” group.

Whitehall Civil Servants Study

A randomized controlled trial was set up to provide an unbiased
estimate of the consequences of smoking cessation in middle-aged
men (50, 51). A total of 1,445 male cigarette smokers with an
especially high cardiorespiratory risk were selected from 16,016 men
aged 40 to 59 who had undergone a cardiorespiratory screening
examination in the Whitehall study of male civil servants in London.
Using a modification of the multivariate linear discriminant func-
tion coefficients that were calculated for predicting coronary heart
disease (CHD) among the Framingham men aged 30 to 62, a risk
score was similarly calculated for each man who smoked five or more
cigarettes a day. This score ranked the smokers according to their
estimated risk of major illness or death from cardiorespiratory
disease. The distribution of the score was tested early in the study,
and a cutoff point was determined such that the scores of 10 percent
of all men and 32 percent of smokers were eligible and exceeded this
value, which was thereafter used to define eligibility for the trial.

Men receiving medical care for heart disease or elevated blood
pressure, those found to have either severe hypertension or diabetes
mellitus, and those with major concomitant disease were excluded
from the trial. Additionally, all men taking psychotropic drugs or
with a record of previous psychiatric inpatient treatment were
eliminated. If during the 1-year interval between initial screening
and trial interview they had died, moved away, or stopped smoking,
they were then not eligible for the study. The remaining 1,445 high
risk eligible cigarette smokers were randomly allocated to study
groups; 714 men composed the intervention group, and 731 men were
in the normal care group.

Men in the intervention group were recalled for a series of
personal interviews with the physician. First, each received a letter
inviting him to an appointment to discuss the results of his previous
examination. At that visit the reason for recall was presented:
evidence in his particular case that smoking represented more than
the average risk to his future health, not the discovery of disease.
The scientific evidence that stopping smoking was likely to bring
benefits was explained and illustrated by charts, with the emphasis

308
 

 

 

 

> 20 { {
<
3 f
oO Oureen
ub
FE
5 mmenQees Normal care
a iP
g antes =intervention
5G

0 ] 4 A i. i A, A. 4. A.

0 5 10

YEARS OF FOLLOWUP

FIGURE 2.—Mean daily cigarette consumption by year of
followup for the Whitehall intervention study
SOURCE: Rose et al. (52).
throughout on the evidence for the positive benefits and practicali-
ties of stopping rather than on the hazards of continuing to smoke.
A full report of the screening examination results and information
that further action was in his hands was sent to the practitioners of
men in the normal care groups. The men were not made aware that
they were involved in a trial. At the 1-year and 3-year points in the
study, they were asked to return for an examination to help
research. Examinations were popular because most men saw them as
beneficial checkups. The questionnaire response rates among survi-
vors were 84 percent at 1 year and 83 percent at 9 years. Dropouts
were mainly retirees. The proportion of responders in the interven-
tion group who were not smoking any cigarettes was 63 percent at 1
year, decreasing to 55 percent at 9 years; initially, all of those in the
study were smokers. Figure 2 shows the trends in stated numbers of
cigarettes smoked in the intervention group. After 1 year, consump-
tion in the intervention group was one-quarter of the normal care
group level. By 9 years, the estimate of cigarette consumption for
intervention men was 70 percent of that for the normal care
controls. Over the 10 years, the net apparent reduction in the
intervention group averaged 7.6 cigarettes/day (53 percent), com-
pared with the control level, as shown in Figure 2 and Table 4.
During the 10 years of followup, there were 128 deaths in the
control group compared with about 130 deaths expected from the
age-specific rates for England and Wales in 1974, as shown in Table
9. The fact that all men entering the trial were high risk smokers
should have increased the observed deaths, but this may have been
offset by the “healthy worker” effect in an occupational study group
or by the selection process that excluded very sick men. Deaths were
also close to national levels for coronary heart disease (111 observed,
94 expected), lung cancer (42 observed, 35 expected), and other

309
TABLE 6.—Summary of deaths and intervention/control
differences from coronary heart disease, all
other causes, and all causes in four randomized
controlled trials

 

 

Percentage
Observed deaths in Expected deaths based difference!
Disease intervention group on control group (O-E)/E
Coronary heart disease
MRFIT 115 123.8 71
Whitehall 49 60.6 -19.1
Osio 6 13.3 -55.4
WHO 367 396.4 ~-74
Unweighted total 537 594.1 -9.6 (p=.02)
All other causes
MRFIT 150 135.8 +105
Whitehall 74 64.4 +149
Oslo 10 9.7 +31
WHO 630 627.7 +04
Unweighted total 864 837.6 +3.2
All causes
MRFIT 265 259.6 +21
Whitehall 123 125.0 ~1.6
Oslo 16 23.0 -30.6
WHO 997 1024.1 -2.7
Unweighted total 1401 1431.7 -2.1

 

‘Unless indicated with a p value, differences are not statistically significant (p > 0.05), based on a two-tailed test
for a Poisson variable.

cancers (40 observed, 41 expected). Seventy-two percent of deaths
occurred in a hospital, and in 45 percent there was an autopsy.
Additional data were obtained from the National Cancer Register of
cases histologically confirmed either by biopsy or at autopsy for
deaths from other causes.

Table 5 shows that the 10-year CHD death rate was 8.5 percent (62
deaths) for the normal care group and 6.9 percent (49 deaths) for the
intervention group, a proportionate change of -19 percent (95
percent confidence limit of 43 to +18 percent). Among the 369 men
who entered the trial with evidence of myocardial ischemia (angina
pectoris, history of possible myocardial infarction, or positive electro-
cardiogram) the reduction was -23 percent compared with -11
percent in those without such evidence. The number of deaths from
cardiovascular causes other than CHD was 12 in the normal care
group and 13 in the intervention group.

Mortality from all causes was initially higher (though not signifi-
cantly so) in the intervention group, but during the last 6 years of
the trial, the rates were higher in the normal care group. During the
whole 10 years, 123 intervention men (17.2 percent) died, compared

310
with 128 (17.5 percent) of the normal care group—a proportionate
change of -2 percent (95 percent confidence limits of -22 percent to
+23 percent). Causes of death were also grouped according to
whether or not they were smoking related. The smoking-related
causes included coronary heart disease, chronic bronchitis, and
cancers of the respiratory tract, esophagus, urinary tract, and
pancreas. There were 92 such deaths in the normal care group and
81 in the intervention group, a proportionate change of -9 percent
(95 percent confidence levels of -31 percent to + 20 percent).

The trial was designed to test whether the total reduction in
cardiorespiratory disease among middle-aged men was as large as
that indicated by the observational studies of ex-smokers. Its size was
planned in the expectation that incidence as well as mortality data
would be available; when this proved unattainable, the resultant loss
of power was partly offset by extending the mortality followup to 10
years. .

At 1-year followup, almost two-thirds of the intervention subjects
had given up cigarettes altogether, while others claimed to be
smoking less than before. Unlike the MRFIT study, objective tests of
smoking behavior were not made here. However, the authors felt
that those who reported complete cessation were generally truthful,
while those claiming to have cut down may have been exaggerating.
The reports were based on questionnaires completed at home, with
little external pressure; they were largely consistent over the
ensuing years. The progressively narrowing gap between the two
groups was due mainly to a gradual reduction in smoking by the
normal care men. Although the size of the gap may have been
overestimated, there is no doubt that throughout the earlier years of
the trial it was large.

Over the trial as a whole, the intervention group’s level of total
smoking exposure was estimated as about half that of the normal
care group, implying that the effects of complete cessation might be
substantially more than those observed in the trial. The results for
total mortality represented the approximate balance of a favorable
trend for smoking-related diseases and an adverse trend for non-lung
cancers. After an exhaustive analysis of the data, the Whitehall
investigators think the difference in non-lung-cancer mortality in
this trial was more likely due to chance than to an effect of
intervention. Such evidence as there is for the latter view should be
considered as a hypothesis for further study, not as the basis for
conclusions or for any recommendation to smokers.

Oslo Study

The purpose of the Oslo study was to find out whether the
cessation of smoking and the lowering of high levels of blood lipids by
dietary changes, if maintained for many years, would lead to

311
reduction in the incidence of first attacks of CHD in men aged 40 to
49 (26, 27, 28). All Oslo men aged 40 to 49 were invited for screening
of coronary risk factors during 1972-73, and 65 percent (16,202 men)
attended. From this cohort, healthy normotensive men were selected
for a controlled trial if they had serum cholesterol levels (mean of
two measurements) of 290 to 380 mg/dl, coronary risk scores (based
on cholesterol levels, smoking habits, and blood pressure) in the
upper quartile of the distribution, and systolic blood pressures (mean
of two measurements) below 150 mm Hg. Those selected had normal
ECGs at rest and were free of any cardiovascular disease, chest pain
on exercise, clinica] diabetes mellitus, fasting blood sugar levels
above 135 mg/dl, cancer, disabling disease, psychopathological
disease, and alcoholism.

Men who met the selection criteria were sent a letter explaining
the experimental design of the trial; 97 percent were found willing to
participate. There were no significant differences between the
intervention and control groups for subject factors such as age,
history of CHD symptoms, cigarette consumption and smoking
status, serum cholesterol and triglyceride, systolic blood pressure,
and diet. After the screening examination, two reexaminations were
made before randomization, the first of these when the subjects were
fasting.

Each of the men in the intervention group was individually talked
with for 10 to 15 minutes by the investigator and introduced to the
risk factor concept and the purpose of the study. Anti-smoking
advice was given individually to all smokers in the intervention
group. They were informed that cessation of smoking might be of
special importance for those with high blood lipid levels. In addition,
the dietitian established a diet record for each man and gave
extensive dietary advice based on this record. Other risk factors were
not subjected to intervention. Followup exams were made every 6
months for intervention subjects and every 12 months for controls.

The intervention of advice on smoking and eating habits resulted
in changes in risk variables. Tobacco consumption, expressed as the
number of cigarettes smoked per man per day, fell about 45 percent
more in the intervention group than in the controls, as seen in Table
4. Pipe smoking was included, taking one pack of pipe tobacco
weekly to equal seven cigarettes daily. The percentage of cigarette
smokers fell by only 16 percent more. The data were assessed by a
questionnaire and by the thiocyanate method. The mean difference
in serum cholesterol between the two groups during the 5 years was
13 percent.

As the design of the study was based on CHD evidence, events of
myocardial infarction (MI) plus sudden death were most important.
CHD mortality (fatal MI plus sudden coronary deaths) was 55
percent lower in the intervention group as compared with the

312
TABLE 7.—Comparison of coronary heart disease incidence
rates (r) and cases (n) in two randomized
controlled trials

 

 

 

Intervention group Control group Percentage
n OI rc ne difference’
Oslo: 5-year results
Fatal MI and
sudden coronary deaths .0099 6 0222 14 ~55.4
_ Nonfatal MI 0215 13 0350 22 -38.6
Total CHD incidence 0314 19 0573 3% ~45.2 (p= .03)
WHO: 6-year results
Fatal MI and
sudden coronary deaths 0150 367 0162 355 -14
Nonfatal MI 0195 406 .0203 401 3.9
Total CHD incidence 0318 773 .0331 756 -3.9

* Unless indicated with a p value, differences are not statistically significant (p > 0.05), based on a two-tailed test
for a Poisson variable.

control group, but the difference was not statistically significant
(p>0.05), as shown in Tables 5 and 7. Total CHD incidence, which
included fatal and nonfatal MI and sudden death, was 45 percent
lower in the intervention group than in the control group, and this
difference was significant (p< 0.05), as shown in Table 7. This is the
only trial to date to show a significant reduction in CHD incidence.
All cases of sudden death satisfied the diagnostic criteria for
coronary death except one unexplained sudden death. The total
coronary deaths, which included fatal MI and sudden death, were
lower, but not significantly lower in the intervention group than in
the control group.

An estimate was made of the proportion of the difference in total
CHD incidence between the two groups (36 versus 19 cases) due to
the reduction in cigarette consumption and of the proportion due to
the reduction in serum cholesterol using logistical regression tech-
niques (28). The percentage of the decline in incidence that was due
to reduction in cigarette consumption was estimated to be 26 percent
among smokers only, predicting a difference between groups of 3.9
cases, out of a total difference of 17 CHD cases. Similar procedures
indicated that the percentage due to cholesterol changes was 60
percent for all men, predicting a difference in CHD incidence of 10.1
cases. It seems, therefore, that the change in cigarette consumption
caused about 25 percent of the difference in CHD incidence between
the two groups and that the difference was due mainly to the
reduction in serum cholesterol. The difference between the groups
was 17 cases: 14 can be explained by this analysis, and the other 3
cases may be due to unexplained intervention effects or to chance.

313
Community-Based Intervention Trials
WHO European Collaborative Trial

As set out in the recent WHO report on the prevention of coronary
heart disease (70), effective prevention must involve the population
as a whole: A high incidence in the population reflects a mass
elevation of risk factors, and most cases of the disease occur in the
large number of people with moderate elevations rather than in the
high risk minority. The first issue of public health policy, therefore,
is to know the effect of population-based prevention, using resources
that could reasonably be afforded. Thus, a trial of CHD prevention
has been undertaken by collaborators in Belgium, Italy, Poland,
Spain, and the United Kingdom (68, 69). It involved the random
allocation of 44 factory pairs, employing 63,732 men aged 40 to 59,
either to a cardiovascular screening and health education interven-
tion program or to a control group. Separate analyses have been
prepared for Belgium factories (35, 36) and United Kingdom factories
(52, 53), but these subgroups are not discussed in detail here.

Randomization of individuals in a trial of community health
education is not feasible, so a somewhat unusual design was used
that randomized entire communities, which in this trial were
factories or other large industrial groups. They were arranged in
matched pairs and randomized, one to intervention and the other to
serve as a control. All men aged 40 to 59 in the intervention factories
were invited to complete a questionnaire and take part in a simple
cardiovascular screening examination during working hours; 87
percent responded. On the basis of a multifactorial scoring system
for risk factor levels (age, job activity, cigarettes smoked per day,
systolic blood pressure, and serum cholesterol), 15 percent or more of
the men in each factory were designated as high risk. A general
campaign of risk factor modification was undertaken with posters,
brochures, personal letters, progress charts, and group discussions
and included advice on reducing or stopping cigarette smoking,
losing weight, lowering serum cholesterol, and increasing leisure
physical activity. The high risk men received more individual
attention, in addition to the general educational program, including
a series of personal sessions with the factory or project physician. All
men with mild hypertension were treated with diuretics or other
drugs in the factory or were referred to a personal physician. Drugs
were not employed to lower serum cholesterol. Annually, a random
sample of men was rescreened. All high risk men were seen at every
anniversary in some centers, or in other centers, at particular
anniversaries. All men remaining in employment at the end of the 5-
or 6-year intervention period were offered a final examination.

In the control factories, a 10 percent random sample of men was
screened initially, and the same 10 percent were asked to return
after 2 years, again at 4 years, and then at the final screening at 5 or

314
6 years, when all remaining men in the original cohort, aged 40 to 59
at the start, were offered screening. It was therefore possible to
compare risk factor levels at the start between all screened
intervention men and 10 percent of controls, between a 5 percent
random sample or more of the intervention men and the same 10
percent of the controls at 2 years and 4 years, and between all
remaining men in both sets of factories at the final screening.
Numbers were reduced by death, by leaving employment, and by
nonresponse to the invitation on the screening day. The risk factor
score used in each intervention factory to designate high risk men
was applied to the 10 percent of the screened men in the paired
control factory in order to identify those who would act as high risk
controls for purposes of measuring risk factor change. Incidence and
mortality results are now available (69) from this controlled trial,
involving randomization of 66 factories (49,781 men) in the United
Kingdom, Belgium, Italy, and Poland (Cracow). Results for Poland
(Warsaw) are not yet complete, and the results have been separately
presented for the United Kingdom (53) and Belgium (35). Net
average reductions in risk factors (all subjects) were 8.9 percent for
daily cigarettes, 1.2 percent for serum cholesterol, 0.4 percent for
weight, 2.0 percent for systolic blood pressure, and 11.1 percent for a
combined CHD risk estimate. Greater reductions occurred in high
risk subjects (19.4 percent for the combined CHD risk estimate).

Tables 5 and 7 show that the net overall changes in CHD rates
were —7.4 percent (95 percent confidence interval from -29 to +15
percent) for deaths (722 deaths) and -3.9 percent (95 percent
confidence interval from -10 to +2 percent) for fatal CHD plus
nonfatal myocardial infarction (1,502 cases). Among men aged 40 to
49 the reduction for this end-point was 15 percent; at ages 50 to 59
there was a small net increase. Deaths from all causes after an early
adverse trend showed a -2.7 percent change overall. There were
large differences between centers, ranging from a 5 percent net
increase in CHD for the United Kingdom to a decrease of 24 percent
in Belgium. In Belgium the decrease both in CHD incidence and in
all deaths was significant at the 5 percent level (35). The effect on
CHD in the different centers correlated broadly with their changes
in risk factors. The authors concluded that a reduction in major
coronary risk factors in industrial populations is possible, but it
depends on adequate resources. The results support the hypotheses
that CHD risk in middle-aged men is reversible and that community
intervention can be beneficial; however, additional followup is
necessary to determine if there are statistically significant reduc-
tions in CHD and total mortality commensurate with the risk factor
changes in the entire factory population.

315
North Karelia Study

The purpose of the North Karelia, Finland, project was to provide
a systematic comprehensive community program to reduce the
currently high mortality and morbidity from CHD, primarily by
reducing the known cardiovascular risk factors of smoking, serum
cholesterol, and blood pressure, while promoting early detection,
treatment, and rehabilitation in people with severe CHD (48, 54, 56).
The focus was on middle-aged males; women were included, but are
not discussed here. A further test was made of the feasibility and the
effect of this approach in the control of CHD and other health
problems on a nationwide level.

Baseline data concerning cardiovascular diseases and their main
risk factors in the target community of North Karelia and in the
control area of Kuopio were studied in detail. To do this, a
representative random 6.6 percent sample was drawn for the 1972
population of the two communities by using the national population
register. The sample comprised men and women born during 1913 to
1947 (then aged 25 to 59). About 52 percent of the men in the study
were current smokers, and each smoker consumed an average of 18
cigarettes per day. Their mean levels of serum cholesterol and blood
pressure were above normal, as seen in Table 4.

Owing to the high general level of the known CHD risk factors in
this population, a comprehensive program was integrated into the
health and social services of the community. The program consisted
of (1) information given to the public, especially about the practical
activities directed toward the risk factors in the community, by
means of media (newspapers, radio, leaflets, posters, and stickers)
and at health information education meetings and public campaigns,
schools, and places of work; (2) organization of services by systemati-
cally integrating the program into existing services and creating new
services when necessary, such as special supporting services for
stopping smoking; (3) training personnel for the special practical
tasks of the program; (4) environmental services to support the
desired lifestyle, for example, with regard to smoking restrictions;
and (5) internal information services to Support registers for
hypertension, myocardial infarction, and stroke and to help with
followup surveys.

No differences were found in the prevalence of smoking between
the target area and the control area at the beginning of the study in
1972. By 1977, there was a nonsignificant net reduction of 2.5
percent in the prevalence of smoking among North Karelia men
relative to the control men. But when the reported amount of
smoking was taken into account, a net 9.8 percent reduction was
significant. This was a result of the finding that in 1972 North
Karelian men smoked more than did those in the control area. There
was a highly significant 4.1 percent net reduction of mean serum

316
TABLE 8.—Death rates and relative risk of death in those
MRFIT subjects who quit at 1 year and those
who continued smoking (average further
followup, 6 years)

 

 

 

Smoking status
at year 1 CHD death rates Total death rates

Smoking status No. No. Percent Percent Percent Percent

at entry quit continued quit continued quit continued
By level of smoking

1-29 cigs/day 613 1,827 1.47 2.30 3.43 4.27

>30 cigs/day 752 4,471 0.80 1.92 3.06 4.47
By group status

Special intervention 991 2,842 111 2.04 2.93 4.68

Usual care 374 3,456 1.07 2.03 4.01 4.20
Total 1,365 6,298 1.10 2.03 3.22 4.41

Relative risk of death (quit/continued)

By level of smoking

1-29 cigs/day 0.64 0.80

> 30 cigs/day 0.42 0.68
By group status

Special intervention 0.54 0.63

Usual care 0.53 0.95
Total 0.54 0.73

 

cholesterol concentrations in the North Karelia men. At the start of
the study, the mean serum cholesterol concentrations were higher in
people in North Karelia than in the control area. There was a highly
significant net reduction in systolic blood pressure from baseline for
the North Karelia men. The prevalence of raised values in 1972 was
similar among men in the two areas, and the net reduction in the
prevalence of raised values in North Karelia was substantial and
highly significant.

The estimates of CHD risk showed that in 1972 the North
Karelians had a higher mean risk score than did the population in
the control area. During the followup period, this difference was
reversed among the men and there was a highly significant net
reduction in the estimated CHD risk in North Karelia of 17.4 percent
among the men. However, there were no significant relative
reductions in CHD or total mortality among the North Karelia men
compared with the control men as of 1977. However, a more recent
analysis of mortality trends in Finland suggests that a longer period
of followup may yield some significant relative reductions (595).
Because of the different methodology used, the mortality results

317
TABLE 9.—Observed deaths in MRFIT and Whitehall
studies and expected deaths based on general
population and Framingham death rates

 

 

 

 

Coronary heart disease All causes
Observed Expected Observed Expected
in UC group (US. males) SMR in UC group (U.S. males) SMR
Study (1979) (1979)
MRFIT study
UC group (n=6438)
Year
1 9 ~19 47 17 ~48 35
2 20 ~20 100 31 ~50 62
36 15 ~87 86 171 ~ 223 V7
7 20 ~24 83 41 ~63 65
17 124 ~150 83 260 ~ 384 68
(Framingham) (Framingham)
1-7 124 ~220 56 260 ~515 50
UC smokers (n= 4091) (U.S. males) (U.S. males)
(1979) (1979)
Years 1-7 89 ~95 94 190 ~244 78
Whitehall study (English males) (English males)
NC group (n=731) (1974) (1974)
Years 1~10 62 ~48 129 128 ~130 98

 

from this study have not been summarized in Tables 5 and 6, but the
risk factor changes are shown in Table 4.

Comparison of Results From intervention Trials and
Epidemiologic Studies

Table 10 summarizes the expected reduction in CHD and total
mortality for former smokers who stopped for 1 to 4 years or for 5 to
9 years, based on the three major observational studies in Table 1
(10, 11, 20, 49). This shows that, relative to current smokers, former
smokers who have stopped for 1 to 9 years have a CHD mortality
rate that is 25 percent less and a total mortality rate that is 18
percent less, in approximate terms. The cigarette smokers in the
MRFIT and Whitehall studies have been subjected to smoking
cessation intervention as the sole, or as a major, risk factor change.
As seen in Table 4, smoking reduction in the intervention group
relative to the control group was 31 percent in MRFIT and 54
percent in the Whitehall study, for an average reduction of about 40
percent in proportion of current smokers. This means that the
expected reduction in mortality should be about 40 percent of that
associated with total cessation; in other words, 10 percent for CHD
mortality and 7 percent for total mortality. Combining the observed

318
6Te

TABLE 10.—Comparison of reductions in CHD and total mortality from observational studies of former

 

 

 

 

 

 

smokers and from MRFIT and Whitehall intervention studies
Coronary heart disease Total mortality
British ACS US. British ACS US.
physicians 25-State veterans Average physicians 25-State veterans Average
Current smokers 1.00 1.00 1.00 1,00 1.00 1.00 1.00 1.00
Former smokers, 1-4 yrs 144 613 854 737 701 934 867 834
Former smokers, 5-9 yrs 887 544 873 768 861 173 809 814
Average former smokers, 1-9 yrs 753 824
Percentage reduction, former smokers, 1-9 yrs ~24.7% -17.6%
Percentage reduction, ~40% former smokers -9.9% -7.0%
O1 Ec %difflO-E)/E Or Ec %diff(O-E)/E
MRFIT smokers at S; (Table 7) 86 89.4 -3.7% 201 190.4 +5.6%
Whitehall smokers (Table 5) 49 60.6 -19.1% 123 125.0 -16%
Unweighted total 135 150.0 -10.0% 324 315.4 +2.1%
95% confidence interval -24% 47% B%—» + 15%

 
deaths from these two intervention groups shows a change in CHD
mortality of -10 percent (95 percent CI from -26 percent to +6
percent) and a change in total mortality of +2.7 percent (95 percent
CI from ~9 percent to +14 percent). Thus, these two trials have
yielded a result after 5 to 10 years of followup that is in good
agreement with the observational studies for CHD deaths, but is not
particularly close for total mortality. Because of the fairly large 95
percent confidence interval, the intervention results are consistent
both with the observational studies and with no change at all. It
must be noted that in MRFIT, other risk factor interventions took
place on blood pressure and cholesterol, and interpretation of the
results has raised the real possibility that an adverse effect on
survival was associated with anti-hypertensive efforts in a specific
subgroup of hypertensive patients, a finding which, if true, might
mask a larger beneficial effect of smoking cessation.

As previously noted and as shown in Table 8, an analysis of
survival in MRFIT participants who had been smokers at entry,
demonstrates that those who quit at year 1 of followup had a 46
percent lower CHD mortality and a 27 percent lower all-cause
mortality than those smokers who did not quit at year 1. This is
further powerful evidence consistent with substantial improvement
of survival associated with the cessation of cigarette smoking.

Conclusions

1. In the four intervention trials involving mortality followup of
individual men for 5 to 10 years, the intervention groups had a
combined total of 10 percent fewer CHD deaths than did the
comparable control groups. Differences for other causes of death
or for total deaths were not significant.

2.In these trials, the amount of cigarette smoking has been
reduced 10 to 50 percent more in the intervention group than in
the control group, demonstrating that intervention can alter
smoking behavior.

3. In the two trials involving morbidity followup, the intervention
groups had 4 and 45 percent lower total CHD incidence than did
the respective control groups.

4. The relative reductions in CHD mortality in each of the four
intervention studies involving individual followup are reason-
ably consistent with the reduction in CHD risk factors, and for a
combination of all four studies, the reduction is statistically
significant.

5. Numerous studies have shown that those who quit cigarette
smoking experience a substantial decrease in CHD mortality
and an improvement in life expectancy.

320
6. A number of prospective epidemiological studies indicate that
former cigarette smokers substantially reduce their CHD and
total death rates from that of current smokers.

321
References

(2)

(2)

(3)

(f)

(5)

(6)

(A)

@

(9

(10)

(1D)

(12)

(13)

(14)
(15)

(16)

(7

(18)

322

AUSTRALIAN NATIONAL BLOOD PRESSURE STUDY. The Australian
therapeutic trial in mild hypertension. Report of the Management Commit-
tee. Lancet 1(8181): 1261-1267, June 14, 1980.

BORHANT, N.O. Primary prevention of coronary heart disease: A critique.
American Journal of Cardiology 40(2): 251-259, August 1977.

BOSSE, R., GARVEY, AJ, COSTA, P.T., Jr. Predictors of weight change
following smoking cessation. International Journal of the Addictions 15(7):
969-991, October 1980.

CEDERLOF, R., FRIBERG, L., HRUBEC, Z., LORICH, U. The Relationship 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. Parts I and II. Stockholm, Karolinska Institute, 1975, 91 pp.

COOPERATIVE RANDOMIZED CONTROL TRIAL. Control of moderately
raised blood pressure. Report of a cooperative randomized controlled trial.
British Medical Journal (5877): 434-436, August.25, 1973.

CORONARY DRUG PROJECT RESEARCH GROUP. The Coronary Drug
Project. Design, methods, and baseline results. Circulation 47(3, Supplement
No. I): I-1—I-50, March 1973.

CORONARY DRUG PROJECT RESEARCH GROUP. Clofibrate and niacin in
coronary heart disease. Journal of the American Medical Association 231:
360-381, January 27, 1975.

CORONARY DRUG PROJECT RESEARCH GROUP. Implications of findings
in the Coronary Drug Project for secondary prevention trials in coronary
heart disease. Circulation 636): 1342-1350, June 1981.

DAYTON, S., PEARCE, M.L., HASHIMOTO, S., DIXON, W.J., TOMTY. ASU,
U. A controlled clinical trial of a diet high in unsaturated fat in preventing
complications of atherosclerosis. Circulation 40(1, Supplement No. ID): I-1—
11-63, July 1969.

DOLL, R., HILL, A.B. Mortality in relation to smoking: Ten years’ observa-
tions of British doctors. British Medical Journal 1(5395): 1399-1410, May 30,
1964.

DOLL, R., HILL, A.B. Mortality in relation to smoking: Ten years’ observa-
tions of British doctors. British Medical Journal 1(5396): 1460-1467, June 6,
1964. -

DOLL, R., PETO, R. Mortality in relation to smoking: 20 years’ observations
on male British doctors. British Medical Journal 2(6051): 1525-1536,
December 25, 1976.

EDERER, F., LEREN, P., TURPEINEN, O., FRANTZ, LD., Jr. Cancer among
men on cholesterol-lowering diets. Experience from five clinical trials.
Lancet 27717): 203-206, July 24, 1971.

ENSTROM, J.E. Trends in mortality among California physicians after giving
up smoking: 1950-79. British Medical Journal 286: 1101-1105, April 2, 1983.

FARQUHAR, J.W. The community-based model of life style intervention
trials. American Journal of. Epidemiology 108(2): 103-111, August 1978.

FARQUHAR, J.W., WOOD, P.D., BREITROSE, H., HASKELL, W.L., MEYER,
A.J., MACCOBY, N., ALEXANDER, J.K., BROWN, B.W., Jr., McALISTER,
A.L., NASH, J.D., STERN, M.P. Community education for cardiovascular
health. Lancet 1(8023): 1192-1 195, June 4, 1977.

FRIEDMAN, G.D., PETITTI, D.B., BAWOL, R.D., SIEGELAUB, A.B. Mortali-
ty in cigarette smokers and quitters. Effect of base-line differences. New
England Journal of Medicine 304(23): 1407-1410, June 4, 1981.

GORDON, T., KANNEL, W.B., McGEE, D., DAWBER, T.R. Death and
coronary attacks in men after giving up cigarette smoking. A report from
the Framingham study. Lancet 2(7893): 1345-1348, December 7, 1974.
(19)

(20)

(21)

(22)

(23)

(24)

(25)

(26)

(27)

(28)

(29)

(30)

(38D

(32)

HAMILTON, M., THOMPSON, E.N., WISNIEWSKI, T.K.M. The role of
blood-pressure control in preventing complications of hypertension. Lancet
1(7327): 235-238, February 1, 1964.

HAMMOND, E.C. Smoking in relation to the death rates of one million men
and women. In: Haenszel, W. (Editor). Epidemiological Approaches to the
Study of Cancer and Other Chronic Diseases. National Cancer Institute
Monograph No. 19. U.S. Department of Health, Education, and Welfare,
Public Health Service, National Institutes of Health, National Cancer
Institute, January 1966, pp. 127-204.

HAMMOND, E.C., GARFINKEL, L. Coronary heart disease, stroke, and
aortic aneurysm. Factors in the etiology. Archives of Environmental Health
19(2): 167-182, August 1969.

HAMMOND, E.C., HORN, D. Smoking and death rates—Report on forty-four
months of follow-up of 187,783 men. I. Total mortality. Journal of the
American Medical Association 166(10): 1159-1172, March 8, 1958.

HAMMOND, E.C., HORN, D. Smoking and death rates—Report on forty-four
months of follow-up of 187,783 men. II. Death rates by cause. Journal of the
American Medical Association 166(11): 1294-1308, March 15, 1958.

HELGELAND, A. Treatment of mild hypertension: A five year controlled
drug trial. The Oslo study. American Journal of Medicine 695): 725-732,
November 1980.

HIRAYAMA, T., HAMANO. Y. Smoking and the relationship with rates for
major causes of death—An epidemiological study based upon 13 years of
continued observation of 265,118 adults over 40 at 29 selected health centers
throughout Japan. Eisei no Shibyo 28(4): 1-18, April 1981.

HJERMANN, I. Smoking and diet intervention in healthy coronary high risk
men. Methods and 5-year followup of risk factors in a randomized trial. The
Oslo study. Journal of the Oslo City Hospitals 30(1): 3-17, January 1980.

HJERMANN, |., VELVE BYRE, K., HOLME, I., LEREN, P. Effect of diet and
smoking intervention on the incidence of coronary heart disease. Report
from the Oslo study group of a randomised trial in healthy men. Lancet
2(8259): 1303-1310, December 12, 1981.

HOLME, I. On the separation of the intervention effects of diet and
antismoking advice on the incidence of major coronary events in coronary
high risk men. The Oslo study. Journal of the Oslo City Hospitals 32(3/4):
31-54, March/April 1982.

HUGHES, G.H., HYMOWITZ, N., OCKENE, J.K., SIMON, N., VOGT, T.M.
The Multiple Risk Factor Intervention Trial (MRFIT). V. Intervention on
smoking. Preventive Medicine 10(4): 476-500, July 1981.

HYPERTENSION DETECTION AND FOLLOW-UP PROGRAM COOPERA-
TIVE GROUP. Five-year findings of the Hypertension Detection and Follow-
up Program. I. Reduction in mortality of persons with high blood pressure,
including mild hypertension. Journal of the American Medical Association
242(23): 2562-2571, December 7, 1979.

HYPERTENSION DETECTION AND FOLLOW-UP PROGRAM COOPERA-
TIVE GROUP. Five-year findings of the Hypertension Detection and Follow-
up Program. II. Mortality by race, sex and age. Journal of the American
Medical Association 242(23): 2572-2577, December 7, 1979.

KARN, H.A. The Dorn study of smoking and mortality among U.S. veterans:
Report on 8 and one-half years of observation. In: Haenszel, W. (Editor).
Epidemiological Approaches to the Study of Cancer and Other Chronic
Diseases. National Cancer Institute Monograph No. 19, U.S. Department of
Health, Education, and Welfare, Public Health Service, National Institutes
of Health, National Cancer Institute, January 1966, pp. 1-125.

323
(33)

(34)

(35)

(36)

(37)

(38)

(39)

(40)

(41)

(42)

(43)

(45)

(46)

(47)

324

KLEBBA, AJJ., DOLMAN, A.B. Comparability of mortality statistics for the
seventh and eighth revisions of the International Classification of Diseases,
United States. Vital and Health Statistics. Series 2, Number 66. U.S.
Department of Health, Education, and Welfare, National Center for Health
Statistics, DHEW Publication No. (HRA)76-1340, October 1975, 93 pp.

KLEBBA, A.J., SCOTT, J.H. Estimates of selected comparability ratios based
on dual coding of 1976 death certificates by the eighth and ninth revisions of
the International Classification of Diseases. Monthiy Vital Statistics Report.
Vol. 28, No. 11, Supplement. U.S. Department of Health, Education, and
Welfare, National Center for Health Statistics, DHEW Publication No.
(PHS)80--1120, February 29, 1980, 19 pp.

KORNITZER, M., DeBACKER, G., DRAMAIX, M., KITTEL, F., THILLY, C.,
GRAFFAR, M., VUYLSTEEK, K. Belgium Heart Disease Prevention
Project: Incidence and mortality results. Lancet 1(8333): 1066-1070, May 14,
1983.

KORNITZER, M., DeBACKER, G., DRAMAIX, M., THILLY, C. The Belgian
Heart Disease Prevention Project. Modification of the coronary risk profile
in an industrial population. Circulation 61(1): 18-25, January 1980.

KULLER, L., MEILARN, E., TOWNSEND, M., WEINBERG, G. Control of
cigarette smoking from a medica] perspective. Annual Review of Public
Health 3: 153-178, 1982.

LEREN, P. The Oslo diet-heart study: Eleven-year report. Circulation 42(5):
935-942, November 1970.

MANN, J.I., MARR, J.W. Coronary heart disease prevention: Trials of diets to
control hyperlipidaemia. In: Miller, N.E., Lewis, B., (Editors). Lipoproteins,
Atherosclerosis and Coronary Heart Disease. Volume 1. Amsterdam, Else-
vier, 1981, pp. 197-210.

MEDICAL RESEARCH COUNCIL RESEARCH COMMITTEE. Controlled
trial of soya-bean oil in myocardial infarction. Lancet 2(7570): 693-700,
September 28, 1968.

METROPOLITAN LIFE INSURANCE COMPANY. Effects of selection on
mortality. Statistical Bulletin 52: 9-11, June 1971.

MIETTINEN, M., TURPEINEN ,0., KARVONEN, M.J., ELOSUO, R., PAAV-
ILAINEN, E. Effect of cholesterol-lowering diet on mortality from coronary
heart-disease and other causes. Lancet 2(7782): 835-843, October 21, 1972.

MULTIPLE RISK FACTOR INTERVENTION TRIAL RESEARCH GROUP.
Multipe Risk Factor Intervention Trial. Risk factor changes and mortality
results. Journal of the American Medical Association 248(12): 1465-1477,
September 24, 1982.

NATIONAL CENTER FOR HEALTH STATISTICS. Health, United States,
1982. U.S. Department of Health and Human Services, Public Health
Service, Health Resources Administration, National Center for Health
Statistics, DHHS Publication No. (PHS)83-1232, December 1982, 191 pp.

NATIONAL DIET-HEART STUDY RESEARCH GROUP. The national diet-
heart study final report. Circulation 37(3, Supplement No. D: I-1—]-428,
March 1968.

NAUKKARINEN, V. The Feasibility of Administering Preventive Treatment
for Coronary Heart Disease to Middle-Aged Men. Helsinki, Finland, K.J.
Gummerus Osakeyhtioen Kirjapainossa J yvaeskylaessae, 1981.

NOPPA, H., BENGTSSON , C. Obesity in relation to smoking: A population
study of women in Goteborg, Sweden. Preventive Medicine 4): 534-543,
July 1980.
(49)

(50)

(51)

(52)

(53)

(54)

(56)

(57)

(58)

(59)

(60)

(6D)

(62)

PUSKA, P., TUOMILEHTO, J., SALONEN, J., NEITTAANMAKI L., MAKI,
J., VIRTAMO, J., NISSINEN, A., KOSKELA, K., TAKALO, T. Changes in
coronary risk factors during comprehensive five-year community pro-
gramme to control cardiovascular diseases (North Karelia project). British
Medical Journal 2(6199): 1173-1178, November 10, 1979.

ROGOT, E., MURRAY, J.L. Smoking and causes of death among US.
veterans: 16 years of observation. Public Health Reports 95(3): 213-222,
May-June 1980.

ROSE, G., HAMILTON, P.J.S. A randomised controlled trial of the effect on
middle-aged men of advice to stop smoking. Journal of Epidemiology and
Community Health 32(4): 275-281, December 1978.

ROSE, G., HAMILTON, P.J.S., COLWELL, L., SHIPLEY, M.J. A randomised
controlled trial of anti-smoking advice: 10-year results. Journal of Epide-
miology and Community Health 36(2): 102-108, June 1982.

ROSE, G., HELLER, R.F., PEDOE, H.T., CHRISTIE, D.G.S. Heart disease
prevention project: A randomised controlled trial in industry. British
Medical Journal 1(6216): 747-751, March 15, 1980,

ROSE, G., TUNSTALL-PEDOE, H.D., HELLER, R.F. UK Heart Disease
Prevention Project: Incidence and mortality results. Lancet 1(8333): 1062-
1066, May 14, 1983.

SALONEN, J.T., PUSKA, P., KOTTKE, T.E., TUOMILEHTO, J. Changes in
smoking, serum cholesterol and blood pressure levels during a community-
based cardiovascular disease prevention program—The North Karelia
project. American Journal of Epidemiology 114(1): 81-94, July 1981.

SALONEN, J.T., PUSKA, P., KOTTKE, T.E., TUOMILEHTO, J., NISSINEN,
A. Decline in mortality from coronary heart disease in Finland from 1969 to
1979. British Medical Journal 286(6381): 1857-1860, June 11, 1983.

SALONEN, J.T., PUSKA, P., MUSTANIEMI, H. Changes in morbidity and
mortality during comprehensive community programme to control cardio-
vascular diseases during 1972-7 in North Karelia. British Medical Journal
2(6199): 1178-1183, November 10, 1979.

SELTZER, C.C. Smoking and coronary heart disease: What are we to believe?
American Heart Journal 100(3): 275-280, September 1980.

SHAPIRO, S., WEINBLATT, E., FRANK, C.W., SAGER, R.V. Incidence of
coronary heart disease in a population insured for medical care (HIP).
Myocardial infarction, angina pectoris, and possible myocardial infarction.
American Journal of Public Health 5%Supplement 6): 1-101, June 1969.

TURPEINEN, O., KARVONEN, M.J., PEKKARINEN, M., MIETTINEN, M.,
ELOSUO, R., PAAVILAINEN, E. Dietary prevention of coronary heart
disease—The Finnish mental hospital study. International Journal of
Epidemiology 8: 99-118, 1979.

U.S. DEPARTMENT OF HEALTH, EDUCATION, AND WELFARE. Smoking
and Health: A Report of the Surgeon General. U.S. 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, 1136 pp.

U.S. PUBLIC HEALTH SERVICE. Smoking and Health. Report of the
Advisory Committee to the Surgeon General of the Public Health Service.
U.S. Department of Health, Education, and Welfare, Public Health Service,
Center for Disease Control, PHS Publication No. 1103, 1964, 387 pp.

U.S. PUBLIC HEALTH SERVICE. The Health Consequences of Smoking. A
Report of the Surgeon General: 1971. U.S. Department of Health, Education,
and Welfare, Public Health Service, Health Services and Mental Health
Administration. DHEW Publication No. (HSM)71-7513, 1971, 458 pp.

325