Y  TABLE I.--Intervention, followup, and cessation results for three major community prevention trials

Community trial

Intervention

Control group contact

Followup

Reported cessation rates/
(objective measures)

WHO European
Collaborative Trial:
United Kingdom
(62)

Mass media intervention for
all factory workers

Antismoking clinics for all  (
smokers

High risk smokers (top IO-15%
risk) offered individual
treatment (four 15 min sessions
in year 1 with company
physician)

Random 10% sample invited    Random 5% IG examined
for screening           yearly

The rest of control males
not told of their
participation in the trial

All survivors examined
at end of trial

Same random 10% CG
screened, reexamined at
2 yea*

Treated

Control

Time

12% '
(high risk
smokers)

ho   5 yrs
change)

9% (all
smokers)

7% (non-
high-risk
smokers)
(no objective measures used)

WHO European
Collaborative Trial:
Belgium
(62)

Mass media intervention for
all factory workers

High risk smokers (top 21%
risk) offered counseling and
examination by project
physicians twice per year

Random 10% invited for
screening

Other 90% had resting
ECG only

Random 5% IG examined
yearly

All survivors examined
at end of trial

18.7% *    12.2%   2 Ym
(high risk (high risk
smokers)   smokers)

12.5% (all 12.6% (all
smokers)  smokers)

(no objective measures used)


TABLE 4.-Continued.

Community trial

Intervention

Control group contact            Followup

Reported cessation rates/
(objective measures)

North Karelia
(Finland) Project
(52, 53, 63)

Comprehensive "community action"
program against risk factors


All forms of media used

Special groups set up aa needed

6.6% random sample examined
at baseline and 5 years

Random 6.6% of each
community surveyed and
examined in 1977

Results compared to
assess RF change

1763     15%     5 Yn
(male     (male
smokers)   smokers)


(SCN drawn for random
    subsamples)
Gxrelations reported. but
adjustments not made)

1P <O.Ol.
'P <0.05.
*Not significant


intervention subjects in the top 21 percent of the risk score
distribution were placed in the high risk group (n= 1,601).

All smokers in the intervention factories received the mass media
approach previously noted for the WHO trials, and all family and
factory physicians received regular information about the partici-
pants' risk factors and took part in intervention (8, 26, 27). Twice a
year, high risk subjects were individually counseled and examined
by two project physicians (8, 26,27) (Table 3).

Reports of the smoking results for this trial have included
comparisons of the results for the 5 percent random sample in the
intervention group with the results for the 10 percent random
sample of the control group at 2-year followup and comparisons of
the results of the high risk subjects in the intervention group with
the results of the high risk subjects selected from the 10 percent
random sample of the control group screened at baseline (9, 26, 27).

Reported smoking rates have not been validated with objective
measures. Among the high risk smokers, 18.7 percent in the
intervention group and 12.2 percent in the control group reported
cessation at 2 years, producing a statistically significant difference
(p ~0.05). For the random samples there was no difference in
reported cessation, with approximately 12.5 percent of the smokers
reporting cessation in both groups (26, 27). Smoking cessation rates
for the intervention group at 1 year was 12 percent and 8 percent for
the high risk subjects and the random sample, respectively (27),
indicating that cessation occurred gradually over the 2-year period.

The North Karelia (Finland) Project

The North Karelia project was carried out in Finland during 1972-
1977 as a comprehensive community program to study the control of
cardiovascular disease (CVD), with special emphasis on CHD, by
reduction of the major alterable CHD risk factors (smoking, in-
creased serum cholesterol, and hypertension (52, 53, 54, 63, 64, 65).
The intervention area was the county of North Karelia in eastern
Finland, which had the highest rates of CVD in that country. The
county of Kuopio, also in eastern Finland, was selected as the control
area because of its similarity to North Karelia.

Both in 1972 and in 1977 a representative random 6.6 percent
sample of the population born between 1913 and 1947 (aged 25 to 59
in 1972 and 30 to 74 in 1977) was drawn from the two counties by
using the national population register (53, 54). The samples in 1972
and in 1977 were independent of each other. Those persons surveyed
were sent a letter explaining the study, a questionnaire assessing
medical history, health behavior and attitudes, attempts to change
health behavior, and stress and an invitation and date for a physical
examination. Over 10,000 subjects were studied each time, with a
participation rate of about 90 percent (52,53,54,63,64,65).


The comprehensive "community action" program against risk
factors was integrated into the health and social services of the
county and was aimed at primary and secondary prevention,
although primary prevention was emphasized. Public information
was provided through newspapers, radio, leaflets, posters, health
education meeetings, and campaigns at schools and places of work
(53); new services were set up if needed, personnel were trained, and
environmental changes (e.g., smoking restrictions) were implement-
ed. The project team planned the activities, prepared the educational
material, helped train personnel, and got the community into action
(54). Smoking cessation group activities were available to those
smokers who wanted them, on the basis of a 3-week model developed
by the project (63). Approximately 55 percent of the smokers were
willing to join the groups, and 71 percent of those who started
completed the groups (63). Approximately 27 percent of those who
started the group reported smoking cessation at 6 months.
The outcomes concerning changes in smoking are based on the
comparison of data obtained in the baseline survey and in the s-year
terminal survey from the study community and the matched control
community. The validity of. the self-reports of smoking behavior was
tested on a random subsample of subjects who were given a second
interview about smoking by trained nurses unaware of the answers
to the survey questionnaire (53). When classified by an interval of 5
or 10 cigarettes, the agreement between the two results was 93 and
97 percent, respectively. The agreement was 99 percent when
classification was smoker or nonsmoker (53). Serum thiocyanate
(SCN) determinations made during the termination interview pro-
vided further validation. Since it is not otherwise noted in the
scientific reports, it is assumed that the results are based on self-
report and are not corrected, Individuals who reported ever having
smoked regularly or having smoked during the preceding month on
an average of more than once a day were classified as smokers. The
reported number of cigarettes, cigars, and pipes smoked per day was
calculated as the amount smoked (53).

The prevalence of smoking in the study and in the control area for
men was 50.2 and 59.9 percent, respectively, and for women 11.7 and
13.1 percent at the start of the study. At year 5, 17 percent of the
baseline male smokers in North Karelia reported smoking cessation
and 15 percent of the baseline male smokers in Kuopio reported
cessation of smoking (52, 53) (Table 4). Thus, smoking had decreased
considerably in both the control and the study groups, yielding a
nonsignificant net reduction in North Karelia of 2.5 percent for the
men and 6.1 percent for the women. With regard to amount smoked,
North Karelian men smoked more than did the men in the control
area in 1972 (9.9 versus 8.9 cigarettes per day), and both groups were
smoking 8.1 cigarettes per day by the end of the study (531, producing

279


a significant net reduction among North Karelian men of 9.8
percent. The mean number of cigarettes smoked by smokers in
North Karelia was 19 cigarettes per day, which remained stable
during the study (63).

A small net reduction occurred in the prevalence of smoking
because, even though considerable cessation was reported in North
Karelia, smoking also decreased at a similar rate in Kuopio, the
control area. The investigators noted several possible explanations
for this decrease (53). There was an increase in interest in antismok-
ing activities toward the end of the study period: the Finnish
Parliament passed new antismoking legislation in 1976, and a new
medical school opened in Kuopio in 1972. They also indicated on the
basis of internal followup surveys that most of the reduction in
smoking occurred at the beginning of intervention in North Karelia,
after the first intensive public antismoking campaign, and that this
lower level of smoking was maintained during the rest of the period
(39. 63).

Deficiencies in the Community Preventive Trials

The major deficiencies in these community preventive trials are
the same as those noted for the clinical trials; i.e., lack of objective
data to verify self-reported outcomes, use of cross-sectional analyses
to the almost complete exclusion of cohort analyses, failure to
provide sufficient information in scientific reports to allow adequate
interpretation of outcomes, and lack of evaluation of components of
the intervention packages.

Objective data to verify self-reports were not used in the United
Kingdom and the Belgium heart disease prevention projects, and
although SCN was measured in the North Karelia study, it was not
used to adjust the self-reported cessation data. Data for strata of
smokers by age were presented for the North Karelia study (74), but
not for cohorts of smokers by smoking-behavior-change categories.
Longitudinal data for cohorts of smokers in the other two trials were
not presented. The value of cohort data is illustrated by a statement
made by the North Karelia investigators in which they noted that
even though the smoking cessation rates were similar in North
Karelia and Kuopio, most of the cessation in North Karelia occurred
at the beginning of the project and was maintained (63, 74). This
information was obtained by the use of followup surveys of samples
of residents. It was hypothesized that most of the cessation for the
comparison community may have occurred near the termination of
the project when antismoking legislation and other changes had
occurred there (39, 63). Cohort data for the comparison community
or for subgroups are not available; thus, the hypothesis cannot be
tested.


Use of the same subjects for baseline and termination surveys are
likely to influence outcomes; therefore, the change may look better
than it actually is (53, 54, 63). The use of a cohort design might
therefore produce a net effect that is not totally a consequence of the
intervention alone, but may also include the effect of the first survey
as well as its interaction with the intervention (63). Thus, there is a
possible need to examine independent cross-sectional population
samples in the two areas under study at the start and termination of
the project. This hypothesis is not supported by data from the North
Karelia study, where cessation rates for 6.6 percent of the random
sample of smokers in the baseline survey of the control community
who were also included in the termination survey were similar to the
rates for the rest of the smokers who were surveyed only at
termination (63). Data from the United Kingdom heart disease
prevention project (62) also failed to support the hypothesis of a
possible intervention effect from screening.

Different components of intervention were not differentially
evaluated within any of the community trials because of the
community orientation of the projects. Thus, conclusions about the
relative contributions of different programs, subprograms, or chan-
nels of action cannot be drawn (39).

Comparison of Community Trials Outcomes

As is noted in Table 4, the 2-year cessation data for both the
Belgium and the United Kingdom WHO trials demonstrate that
there were significant differences in reported cessation rates for the
intensively intervened-with high risk smokers as compared with the
smokers in the control factories, but there were no significant
differences between the non-high-risk smokers in the intervention
factories who received a media-only approach as compared with the
non-high-risk smokers in the control factories. This outcome is
similar to the previously noted finding in the Stanford study; i.e.,
media only had no more intervention impact than had no interven-
tion. This lack of demonstrated impact for a media approach to
smoking cessation in the Belgium study was in part due to the 12.5
percent cessation rate achieved by the control group. This occur-
rence of cessation in a control group is similar to that demonstrated
in each of the clinical trials. Again, a saturation point may have been
reached in groups in which there is already an increased level of
awareness, and intensive intervention may be necessary if additional
cessation is to be realized in the next level of smokers. It may also be,
as previously noted, that although cessation occurred among the
nonintervention smokers, the long-term maintenance rate among
those in this group who stop smoking may be significantly different
from the long-term maintenance rate for the intervention group


smokers. Because of the lack of cohort data, this issue cannot be
addressed.

The range of cessation rates among the comparison groups for the
three trials is large, 0 to 15 percent, with the lowest rate recorded for
the United Kingdom group of the WHO collaborative study and the
highest for the Belgium group of the same study. The 0 percent
cessation rate for the control group in the United Kingdom trial is
puzzling, as this is less than the spontaneous cessation rate observed
in the general population. More in-depth analysis of the data for the
10 percent random sample of the control group who were screened at
baseline and reexamined at 2 years is indicated. Different protocols
for contact with the control group were used for the United Kingdom
from those used for the Belgium groups. The most notable difference
was that 90 percent of the United Kingdom control group were not
told of their participation in the trial, but 90 percent of the Belgium
group were told of their participation and had a resting ECG. An
additional 10 percent were told of their participation and had
complete physical examinations. It can be hypothesized that the use
of an ECG may have had an intervention effect for the Belgium
control group. The 15 percent rate for the North Karelia study was
determined at 5 years. One might hypothesixe that this rate would
be lower at 2 years, the point at which the other two studies
conducted followup.

The 5year cessation rate for the United Kingdom collaborative
trial is also low when compared with the 2-year rate for the Belgium
collaborative trial, which utilized a similar protocol. A major
difference for these two groups was that high risk smokers in the
Belgium study received two examinations per year while those in the
United Kingdom were given one. There was also a difference in the
number of physician-intervention visits during year 1; two were used
in the Belgium study and four were used in the United Kingdom
trial. Perhaps the feedback provided by an examination has a
greater intervention effect than a session with a physician intended
for counseling only. It can also be hypothesized that cultural
differences may have affected the differences in outcome between
the two groups in the same trial.

In general, the use of community programs that used only a media
approach did not produce a greater intervention effect than was
observed in the comparison community. The incorporation of more
intensive intervention in groups in addition to the media approach
was necessary before significant differences could be realized. The
same outcomes were observed in the Stanford study.

282


Conclusions
1. Smokers involved in intervention programs demonstrate higher
  smoking cessation rates than those in control groups.
2. In general, the success of smoking intervention programs is
  related to the amount of intervention provided.


References

  (II BARRETT, T.J., SACHS, L.B. Test of the classical conditioning explanation of
    covert sensitization. Psychological Reports 34(3): 13121314, June 1974.
  (2) BENFARI, R.C. The Multiple Risk Factor Intervention Trial MRFIT) III: The
    model for intervention. Preventive Medicine lo(4): 4M2, July 1981.
(3) BENFARI, R.C., MCINTYRE, K., BENFARI, M.J.F., BALDWIN, A., OCK-
   ENE, J. The use of thiocyanate determination for indication of cigarette
    smoking status. Evaluation Quarterly l(4): 629638, November 1977.
  (4) BENOWITZ, N.L., HALL, S.M., HERNING, R.I., JACOB, P. III, JONES, R.T.,
    OSMAN, A.-L. Smokers of low-yield cigarettes do not consume lees nicotine.
    New England Journ of Medicine 309(3): 139-142, July 21,1983.
(5) BERNSTEIN, D.S. Modification of smoking behavior: An evaluative review.
    Psychological Bulktin 71(6): 41E?440, June 1969.
  (61 CHEN, T. Smoking Behuvioral Change and Its Correlates in a Selected Group
    of Middle-Aged Men. Doctoral dissertation, Ohio State University, Universi-
    ty Microfilms International, Ann Arbor, Michigan, Thesis number 71-7422,
     1970.
  (7) COLTON, T. Statistics in Medicine. Boston, Little, Brown and Company, 1974,
    372 pp.
(8) DeBACKER, G., KORNITZER, M., THILLY, C., DEPOORTER, A.M. The
    Belgian multifactor preventive trial in CVD (I) design and methodology.
    Hart Bulletin 6(6): 143-146, December 1977.
  (9) DerSIMONIAN, R., CHARETTE, L.J., McPEEK, B., MOSTELLER, F. Report-
    ing on methods in clinical trials. New Engkmd Journal of Medicine 306f22):
     1332-1337, June 3,1982.
(10) DiCLEMENTE, C.C., PROCHASKA, J.O. Self&ange and therapy change of
    smoking behavior: A comparison of processes of change in cessation and
    maintenance. Addictive Behaviors 7(2): 13%142,1982.
(II) EVANS, RI., HANSEN, W.B., MI'ITLEMARK, M.B. Increasing the validity
    of calf-reports of smoking behavior in children. Journal of Applied Wycholo-
    gy 62(4): 521523, April 1977.
(12) EVANS, D., LANE, D.S. Long-term outcome of smoking cessation workshops.
    American Journal of Public Health 70(7): 725727, July 1989.
(13) FARQUHAR, J.W. The community-based model of life style intervention
   trials. American journal of Epidemiology 108(2): 103-111, August 1978.
(Zn 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 l(8023): 1192-1195, June 4,1977.
(Z5l GRITZ, E.R. Smoking behavior and tobacco abuse. In: Mello, N.K. (Editor).
    Advances in Substance Abuse. Volume 1. Greenwich, Connecticut, JAI
    Press, 1986, pp. 91-158.
(16, HJERMANN, I. Smoking and diet intervention in healthy coronary high risk
    men. Methods and 5year follow-up of risk factors in a randomized trial. The
    Oslo study. Journal of the Qslo Hospitals 30(l): 3-17, January 1989.
(17) HJERMANN, I., VELVE BYRE, K., HOLME, I., LEREN, P. Effect of diet and
    smoking intervention on the incidence of coronary heart diaeaae. Report
    from the Oslo study group of a randomized trial in healthy men. Lancet
    2(8259): 1303-1310, December 12,198l.
(I81 HOLME, I., HELGELAND, A., HJERMANN, I., LEREN, P. The Oslo study:
    Social indicators, risk factors and mortality. In: Bostrom, H., Ljungstedt, N.
    (Editors). Medical Aspects of Mortality Statistics. Skandia International
    Symposia. Almquist and W&sell International, Stockholm, Sweden, 1981,
    pp. 165-181.


284


(19) 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 lO(4): 476-500, July 1981.
(20) HUNT, W.A., BESPALEC, D.A. An evaluation of current methods of
   modifying smoking behavior. Journal of Clinical Psychology 30(4): 431-438,
    October 1973.
(21) JACOB, P., III, WILSON, M., BENOWITZ, N.L. Improved gas chromatograph-
   ic method for the determination of nicotine and cotinine in biologic fluids.
   Journ& of Chromatogruphy 222(l): 61-70, January 2.1981.
(22) JACOBS, M.A., SPILKEN, A.Z., NORMAN, M.M., WOHLBERG, G.W.,
   KNAPP, P.H. Interaction of personality and treatment conditions associ-
   ated with success in a smoking control program. Psychosomatic Medicine
   33(6): 545-556, November-December 197 1.
(23) KANZLER, M., ZEIDENBERG, P., JAFFE, J.H. Response of medical person-
   nel to an on-site smoking cessation program. Journal of Clinical Psychology
   32(3): 670-674, July 1976.
(24) KASL, S.V. Cardiovascular risk reduction in a community setting: Some
   comments. Journal of Consulting and Clinical Psychology 43(2): 143-149,
   April 1980.
(25) KEUTZER, C.S. Behavior modification of smoking: The experimental investi-
   gation of diverse techniques. Behavior Research and Therapy 6(2): 137-157,
   May 1968.
(26 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(l): 18-25, January 1980.
(27) KORNITZER, M., DRAMAIK, M., KI'ITEL, F., DeBACKER, G. The Belgian
   Heart Disease Prevention Project: Changes in smoking habits after two
   years of intervention. Preventive Medicine 9(4): 4QW503, July 1980.
(28) LANDO, H.A. A comparison of excessive and rapid smoking in the moditica-
   tion of chronic smoking behavior. Journal of Consulting and Clinical
   Psychology 43(3): 350355, June 1975.
(29) LANDO, H.A. Measurement and technique innovations. An objective check
   upon self-reported smoking levels: A preliminary report. Behavior Thempy
   q4): 547-549, July 1975.
(30) LANDO, H.A., McGOVERN, P.G. Three-year data on a behavioral treatment
   for smoking: A follow-up note. Addictive Behaviors 7(2): 77-181,1982.
(31) LANGONE, J.J., Van VUNAKIS, H., HILL, P. Quantitation of cotinine in
   sera of smokers. Research Communications in Chemical Pathology and
   Pharmacology. lo(l): 21-28, January 1975.
(32) LEVENBERG, S.B., WAGNER, M.K. Smoking cessation: Long-term irrele-
   vance of mode of treatment. Journal of Behavior Therapy and Experimental
   Pbychiutry 7(l): 93-95, March 1976.
(33) LEVENTHAL, H. Experimental studies of anti-smoking communications. In:
   Borgatta, E.F., Evans, R.R. (Editors). Smoking, Health, and Behavior.
   Chicago, Aldine Publishing, 1968, pp. 95-121.
(34) LICHTENSTEIN, E., DANAHER, B.G. Modification of smoking behavior: A
   critical analysis of theory, research, and practice. Advances in Behavior
   Modifmtion 3: 79-132,1976.
(35) LUEPKER, R.V., PECHACEK, T.F., MURRAY, D.M., JOHNSON, C.A.,
   HUND, F., JACOBS, D.R. Saliva thiocyanate: A chemical indicator of
   cigarette smoking in adolescents. American Journal of Public Health 71(12):
   1320-1324, December, 1981.


(36) MACCOBY, N., FARQUHAR, J.W., WOOD, P.D., ALEXANDER, J. Reducing
  the risk of cardiovascular disease: Effects of a community-based campaign
  on knowledge and behavior. Journal of Community Health 3(2): 10&114,
   Winter 1977.

(37) MARSTON, A.R., &FALL, R.M. Comparison of behavior modification
  approaches to smoking reduction. Journul of Consulting and Clinical
   parChology 36(2): 153-162, April 1971.

(38) MAlTSON, M., FURBERG, C.D., DUNBAR, J., SCHUCKER, B. Behavioral
  sciences in clinical trials An overview. Behavioral Medicine Update 2(d): b
   10, Fall 1982.

(39) McALISTER, A.. PUSKA, P., SALONEN, J.T., TUOMILEHTO, J., KOSKE
  LA, K. Theory and action for health promotion: Illustrations from the North
  Karelia Project. American Journal of Public Health 72(l): 43-50, January
    1982.

(40) McFALL, R.M. HAMMEN, CL. Motivation, structure, and self-monitoring:
  Role of nonspecific factors in smoking reduction. Journul of Consulting and
   Clinical pSychologv 37(l): 80-86, August 1971.

(41) MEYER, A.J., NASH, J.D., McALISTER, A.L., MACCOBY, N., FARQUHAR,
  J.W. Skills training in a cardiovascular health education campaign. Journal
  of Consulting and Clinical Psychology 48(2): 12%142, April 1980.

(42) MIlTLEMARK, M.B. Information on Imminent Versus Long Term Health
  Consequences: Impact on Childmn's Smoking Behavior, Intentions, and
  Knowledge. Doctoral dissertation, University of Houston, University Micro-
  films International, Ann Arbor, Michigan, Thesis number 78-18238, May
   1978,166 pp.

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

(4 NASH, J.D.. FARQUHAR, J.W. Applications of behavioral medicine to
  disease prevention in a total community setting: A review of the three
  community study. In: Fergueon, J.M., Taylor, C.B. (Editors). The Comprehen-
  sive Handbook of Behavioml Medicine. Volume 3: Extended Applications
  and Issues. Spectrum Publications, Jamaica, New York, 1980, pp. 313-335.
(45) NATIONAL INTERAGENCY COUNCIL ON SMOKING AND HEALTH.
  Guidelines for Research on the Effectiveness of Smoking CeaPation Programs.
  A Committee Report. National Interagency Council on Smoking and Health,
  American Dental Association, Chicago, October 1974,46 pp.

(46) NEATON, J.D., BROSTE, S., COHEN, L., FISHMAN, E.L., KJELSBERG,
  M.O., SHOENBERGER, J. The Multiple Risk Factor Intervention Trial
  (MRFIT). VII. A comparison of risk factor changes between the two study
   groups. Preventive Medicine lO(4): 519-543, July 1981.
(47) OCKENE, J.K. A Study of the Psychosocial Factors Involved in Changing
  Smoking Behavior: Risk Factor Altemtion in a Coronary Heart Disease
  Prevention Rugrum Doctoral Thesis, Boston College, Department of Educa-
  tion, Boston. University Microfilms International, Ann Arbor, Michigan,
   Thesis number 79-22078,1979,286 pp.

(48) OCKENE, J.K., BROSTE, S., HYMOWlTZ, N., HUGHES, G., LAUGER, G.
  For the MRFIT Research Group Paper presented at Council on EpidemioI*
  gy, American Heart Association, San Diego, California, March 1983.
(49) OCKENE, J.K., HYMOWITZ, N., SEX'ION, M., BROSTE, SK. Comparison of
  patterns of smoking behavior change among smokers in the Multiple Risk
  Factor Intervention Trial 0. Preventive Medicine ll(6): 621-638,
   November 1982.

286


(50) OCKENE, J.K., NUTALL, R., BENFARI, R.C., HURWITZ, I., OCKENE, IS. A
   psychosocial model of smoking cessation and maintenance of cessation.
   Fhmentiw Medicine lO(5): 623638, September 1981.
(51) PEITIGREW, A.R., FELL, G.S. Microdiffusion method for estimation of
  cyanide in whole blood and its application to the study of conversion of
   cyanide to thiocyanate. Clinical Chemistry 19(5): 466-471,1973.
(52) PUSKA, P., KOSKELA, K., PAKARINEN, H., PUUMALAINEN, P., SOINI-
  NEN, V., TUOMILEHTO, J. The North Karelia Project: A programme for
   community control of cardiovascular diseases. Scandinavian Journal of
   Social Medicine 4(2): 57-60.1976.
(53) PUSKA, P., TUOMILEHTG, T., SALONEN, J., NEITTAANMAKI, L., MAKI,
  J., VIRTAMO, J., NISSINEN, A., KOSKELA, K., TAKALG, T. Changes in
   coronary risk factors during comprehensive five-year community pro-
  gramme to control cardiovascular diseases (North Karelia Project). British
   Medical Joumul2(6198): 1173-1178, November 10,1979.
(54) PUSKA, P., VIENOLA, P., KOlTKE, T.E., SALGNEN, J.T., NEI'ITAAN-
  MAKI, L. Health knowledge and community prevention of coronary heart
  disease. Internntional Journal of Health Educatioo 24(Supplement 2): 1-11,
   AprilJune 1981.

(55) REID, D.D., HAMILTON, P.J.S., KEEN, H., BRETT, G.Z., JARRE'IT. R.J.,
  ROSE, G. Cardiorespiratory diseases and diabetes among middleaged male
   civil servants. Luncet l(7856): 469-473, March 23,1974.
(56) RINGOLD, A., GOLDSMITH, J.R., HELWIG, H.L., FINN, R., SCHUEITE, F.
  Estimating recent carbon monoxide exposures. A rapid method. Archives of
   Environmental Health 5(4): 308-318, October 1962.
(57) RONAN, G., RUANE, P., GRAHAM, I.M., HICKEY, N., MULCAI-IY, R. The
   reliability of smoking history amongst survivors of myocardial infarction.
   British Joumnl of Addiction 76(4): 425-428, December 1981.
(58) ROSE, G. A controlled trial of the effects of health education in high-risk
   subjects. Ann&s o!es So&tea Be&es de Medecine Tzolpicale 50(4): 4814,
    1970.

(59) ROSE, G. Physician counseling and personal intervention. In: Steinfeld, J.,
   Griffiths, W., Bail, K., Taylor, R.M. (Editors). Health Consequences, Educa-
   tion, &sea&n Activities, and Governmental Action. Volume II. Proceedings
  of the Third World Conference on Smoking and Health, New York City,
   June 2-5, 1975. U.S. Department of Health, Education, and Welfare, Public
   Health Service, National Institutes of Health, National Cancer Institute,
   DHEW Publication No. (NIH)77-1413,1977, pp. 515-523.
(60) ROSE, G., HAMILTON, P.J.S. A random&d controlled trial of the effect on
  middle-aged men of advice to stop smoking. Journal of Epidemiology and
   Community HeaZth 32(4): 275-281, December 1978.
(61) ROSE, G., HAMILTON, P.J.S., COLWELL, L., SHIPLEY, M.J. A random&d
  controlled trial of anti-smoking advice: X&year results, Journal of Egio!e-
   miology and Community Health 36(2): 102-108, June 1982.
(62) ROSE, G., HELLER, R.F., PEDOE, H.T., CHRISTIE, D.G.S. Heart disease
   prevention project: A random&d controlled trial in indstry. British Medical
   Joumul l(6216): 747-751, March 15,198O.

(63) SALGNEN, J.T., HEINONEN, O.P., KO'ITKE, T.E., PUSKA. P. Change in
  health behaviour in relation to estimated coronary heart disease risk during
   a community-based cardiovascular disease prevention programme. Znternu-
   tionnl Journal of Epidemiology lO(4): 343-354, December 1981.
(64) SALONEN, J.T., PUSKA. P., KOTI'KE, T.E. Smoking, blood pressure and
  serum cholesterol as risk factors of acute myocardial infarction and death
   among men in eastern Finland. European Heart Journal U5): 365-373,1981.

287


(65J SALONEN, J.T., PUSKA, P., KO'ITKE, T.E., TUOMILEHTO, J. Changes in
  smoking, serum cholesterol and blood pressure 1eveIs during a community-
  based cardiovascular disease prevention program-the North Karelia
  Project. American Journal of Epidemiology 114(l): 81-94, July 1981.
(66) SCHACHTER, S. Recidivism and selfcure. of smoking and obesity. American
   &cholog& 37(4): 436-444, April 1982.
(67) SCHWARTZ, J.L. A critical review and evaluation of smoking control
  methods. Public Health Reports &L(6): 489566, June 1969.
(68) SCHWARTZ, J.L. Smoke Watchem Evalwtion &liminary Report) Unpubl-
  ished report, Institute of Health Research, Berkeley, 1973.
(691 SCHWARTZ, J.L. Research methodology in smoking cessation: A critique. In:
  Steinfeld, J., GriffXhs, W., Ball, K., Taylor, R.M. (Editors). Health Conse-
  quences, Education, Cessation Activities, and Governmental Action. Volume
  II. Proceedii of the Third World Conference on Smoking and Health, New
  York City, June 2-5, 1975. U.S. Department of Health, Education, and
  Welfare, Public Health Service, National Institutes of Health, National
  Cancer Institute, DHEW Publication No. (NIH)77-1413, 1977, pp. 649653.
(70) SCHWARTZ, J.L., DUBITZKY, M. PSychasocial Factors ZnvoZved in Cigarette
  Smoking Cessation. Berkeley, Institute of Health Research, Permanente
   Medical Group, Kaiser Foundation Health Plan, 1968.
(71) SHIPLEY, R.H., ROSEN, T.J., w, C. Measurement of smoking:
  Surveys and some recommendations. Addictive Behaviors 7(3): 299-302,
    1982.
(72) STEWART, R.D. The effect of carbon monoxide on humans. Annual Review of
   Pharmacology 15: 409-423,1975.
(73) STRAITS, B.C. The discontinuation of cigarette smoking: A multiple discrimi-
  nant analysis. In: Zagona, S.V. (Editor). Studies and Issues in Smoking
  Behavior. University of Arizona Press, Tucson, 1967, pp. 79-81.
(74) TUOMILEHTO, J., KOSKELA, K., PUSKA, P., BJORKQVIST, S., SALG
  NEN, J. A community anti-smoking programme: Interim evaluation of the
  North Karelia Project. Znternational Journul of Health Education 21(4,
   Supplement): 3-15, October-December 1978.
(75) 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, Oftice of the Assistant
  Secretary for Health, Office on Smoking and Health, DHEW Publication No.
   (PHS)7950066,1979,1136 pp.
(76) WALD. N.J., IDLE, M., BOREHAM, J., BAILEY, A. Carbon monoxide in
  breath in relation to smoking and carboxyhaemoglobin levels. 2'hom.x 36(5):
   366-369, May 1981.
(77) WALLACE, N.D., DAVIS. G.L., RUTLEDGE, R.B., KAHN, A. Smoking and
  carboxyhemoglobin in the St. Louis metropolitan population. Archives of
   Environmental Health 29(3): 136-142, September 1974.
(78) WERKO, L. Prevention of heart attacks. A multifactorial preventive trial in
  Gothenburg, Sweden. Ann& of Clinical Research ll(2): 71-79, April 1979.
(79) WILHELMSEN, L. Risk factors for disease according to population studies in
  G<iteborg, Sweden. Paper presented at Skandia International Symposia,
  September 23-25, 1980. In: Bostrom, H., Ljungstedt, H. (Editors). Medical
  Aspects of Mortality Statistics. Almquist and W&sell International, Stock-
   holm, Sweden, 1981, pp. 73-88.
(80) WILHELMSEN, L., TIBBLIN, G., WERKO, L. A primary preventive study in
  Gothenburg, Sweden. Preventive Medicine lfl-2):153-160, March 1972.


(81) WORLD HEALTH ORGANIZATION EUROPEAN OLLABORATIVE
  GROUP. An international controlled trial in the multifactorial prevention
  of coronary heart disease. Inkrnationd Journal of Epidemiology 3(3): 219-
   224, September 1974.

(82) ZEIDENBERG, P., JAFFE, J.H., KANZLER, M., LEVITI', M.D., LANGONE,
  J.J., Van VUNAKIS, H. Nicotine: Cotinine levels in blood during cessation
  of smoking. Comprehensive Psychiaty 18(l): 93-101, January/February
    1977.

289


SECTION 8. THE EFFECT OF
        CIGARETTE SMOKING
         CESSATION ON
         CORONARY HEART
         DISEASE

291


Epidemiologic Evidence Regarding Smoking Cessation and
Coronary Heart Disease

The epidemiologic data on smoking and coronary heart disease
(CHD) were reviewed in detail in a preceding section, as well as in
the Reports of the Surgeon General for 1964,1971, and 1979 (60, 61,
62). Coronary heart disease (ICD/G and ICD/7 No. 420) before 1968
and ischemic heart disease (ICD/8 and ICD/9 Nos. 410414) since
1968 are considered synonymous with one another for all practical
purposes and are abbreviated as CHD. Terminology and data on
CHD are discussed in detail elsewhere (33, 34, 44). This discussion is
limited to epidemiologic data on smoking cessation and CHD.
Several prospective studies involving self-selected questionnaire
respondents include extensive epidemiologic data on smoking cessa-
tion and CHD mortality. The results, summar@ed in Table 1, show
CHD death rates for former smokers relative to never smokers as a
function of the number of years stopped smoking cigarettes, general-
ly determined as of the time the questionnaire was completed. Data
in this form are available only for men, generally white men. The
studies are as follows: the British physicians study, including 10-year
followup (10, II) and 20-year followup (12); the American Cancer
Society S-State study (22, 23) and the American Cancer Society 25
State cancer prevention study (20, 21); the U.S. veterans study with
&&year followup (32) and X-year followup (49); and the Swedish
representative sample study with l@year followup (4). Excluded
were numerous studies that present data only on former smokers as
a whole or have data on a few special categories of former smokers,
such as Shapiro et al. (58) and Hirayama and Hamano (25). Much of
these other epidemiologic data on former smokers is summarized in
the 1979 Report of the Surgeon General on Smoking and Health (60)
and in the preceding section of this Report on coronary heart disease.

Numerous epidemiologic studies (10, II, 12, 14, 20, 21, 32) have
shown a decrease in CHD mortality for ex-smokers compared with
continuing smokers, and it has been suggested that smoking
cessation accounts for this salutary effect. Another view (57) has
been that continuing smokers and quitters are somehow constitu-
tionally different and that their health experiences might also be
different, independent of smoking status. Two prospective studies of
current smokers, some of whom became persistent quitters during
the course of the study, show that persistent quitters have lower
CHD and total death rates than do continuing smokers (17, 18).
Friedman et al. (17) examined this question in some detail in the
Kaiser-Permanente study of over 25,000 persons. They compared 18
baseline characteristics related to coronary disease in quitters and
continuing smokers at a time when all were smoking. They found
that the beneficial effects of quitting on CHD mortality could not be
explained by differences in their baseline characteristics (Table 2).

293


F  TABLE l.-Male coronary heart disease and total mortality ratios for current and former cigarette
       smokers relative to never smokers, as a function of years stopped smoking

Overall cohort        Smoking selection
description                criteria

Years
stopped '

Coronary heart disease
mortality ratbE

All causes
mortality ratio2

British physicians'









British physicians'









American Cancer Society
S&ate study S

Cigarettes only         0                1.41 (464)                        1.37 (1566)
                l-4                1.05 (28)                         0.96 (71)
                5-9                1.25 (61)                         1.18 (204)
               10-14               1.16 (59)                          1.12 G?O4)
               15+               1.12 (40)                         1.11 (153)
                NS               1.00 (113)                        1.00 (436)

                          Attained age                     Attained age

                         3c-54        55-64       65+         30-64        65+        30+

Cigarettes only,        0        3.5         1.7         1.3          2.0         1.6          1.8
at least 5 years        l-4        1.9 (71       1.9 (19)      1.0 (24)       1.7 (67)      1.4 (99)      1.5 (1661
                5-9        1.3 (10)       1.4 (34)      1.3 (76)       1.6 (141)      1.4 (242)      1.5 (383,
               lo-14       1.4 (101      1.7 081      1.2 (62)       1.4 (104)      1.2 (206)      1.3 (310)
              15+       1.3 (7)       1.3 (45)      1.1 (148)      1.1 w6)      1.1 wEa      1.1 (590)
               NS       1.0 (32)      1.0 (75)      1.0 w?2)      1.0 (326)      1.0 (611)      1.0 (9371

                       Cigarettes smoked per day             Cigarettes smoked per day

                             l-19           2t+                 1-19           20+

Cigarettes only         0           1.75 @J4)        2.20 (6041             1.61 (2303)        2.02 (1326)
               <l           2.10 (23)         3.00 (18)              2.04 (51)         2.69 05)
                l-9           1.54 (80)         2.06 (64)              1.30 (159)        1.82 (135)
              lO+           1.09 (401         1.60 (40)              1.08 (141)        1.50 (87)
               NS           1.00 (709)        1.00 (709)              1.00 (1644)        1.00 wx4)


TABLE I.-Continued.

Overall cohort         Smoking selection
description                criteria

Coronary heart disease
mortality ratio'

All causes
mortality ratio'

American Cancer Society
25State study'

Cigarettes only

Cigarettes smoked per day

for initial ages 40-79

              1-19           20+

0           1.90 (1063)        2.55 (2822)
*. 1           1.62 (29)        1.61 (62)
l-4           1.22 (57)         1.51 (154)
5-9          1.26 (55)         1.16 (135)
10-19          0.96 (52)         1.25 (133)


201           1.08 (70)         1.05 (80)

Cigarettes smoked per day
   for initial ages 54S74

1-19           20+

1.72 I20151        1.94 (3741)
1.61 (641         2.18 (213)
1.44 (1441        1.98 (499)
1.34 (1261        1.49 (416)

1.02 (2551        1.32 (5461

NS           1.00 (1841)       1.00 (1841)             1.00 (3512)        1.00 (3512)

Attained age. 55-64

Attained age. 55-64

U.S. veterans'        Cigarettes with or       0                  1.66 (3064)                          1.72 (6928)
              without cigars/pipes,     l-l                 1.34 (155)                           1.56 (379)
             stopped for other       5-9                 1.47 (279)                           1.42 (5961
                than doctor's orders     lo-14                 1.13 (161)                           1.28 0651
                               15+                 0.97 (342)                           107 (7791
                                NS                1.00 (1218)                          1.M) (2617)

                                                 Attained age, 31-99                     Attained age, 31-99

U.S. veterans'        Cigarettes with or       0                  1.56 (13,645)                          1.73 (36,143)
              wthout cigars/pipes,     14                 1.35 uO1                          -1.5  (384)
              stopped for other       5-9                 1.36 (599)                          -1.4  (1441)
               than doctor's orders     lo-14                 1.29 (997)                          - 1.3  (2445)
                                E-19                 1.21 (1101)                         - 1.2  (2767)
                                20+                 1.05 (2418)                         -1.05 (6049)

E
01                                     NS                1.M) (-6500)                      1.00 (16,224)


TABLE L-Continued.

Overall cohort
description

Smoking selection
   criteria

Years
stopped I

Coronary heart disease
mortality ratio*

All causes
mortality ratio '

Swedish representative
samples

0
l-9 '
10+
NS

Years smoked cigarettes, initial age, 40-69      Years smoked cigarettes, initial age, 40-69

<20         >20        Total         <20         >20
         L                                      Total
                                         A
                   1.7 (212)                            1.4 (557)
0.9 (71       1.6 (84J       1.5 (97)       1.0 (26)       1.4 (212)      1.3 (253)
0.9 (40)      1.1 (46)      1.0 (86)       1.0 (1231      1.0 (117)      1.0 (241)
1.0 (219)      1.0 (219)      1.0 (219)      1.0 (671)      1.0 (671)      1.0 (671)

' Years stopped smoking was measured as of beginning of followup. except for the U.S. veterans study. where the number of years stopped was increased by 1 with the passage of each calendar year
unless death occurred 0 years stopped denotes current smoker; NS denotes never smoker.
`Mortality ratio is former emoker death rate relatrve to never smoker death rate, properly adjusti for age; ratio for never smokers is delined to be 1.0. Number of deaths are in parentheses.
"Study of 34,445 men aged 201. at lOyear followup. 1951-1961. Doll end Hill (10. 11).
.Study of 34,440 men aged 20+. et Myear followup. 1951-1971. Doll and Peta (12).
`Study of 187,783 men aged 5O-fj9. et 44.month followup, 1952-1955. Hammond and Horn (22,23).
`Study of 440,558 men aged 301, approximately at 4-yeer followup. 1959-1963. for total mortality, end 358,534 diseasefree men at Gyear followup. 1959-1965, for CHD mortality Hammond (20),
Hammond and Carfinkel(21).

`Study of 248,046 men a& 31-&1. at 5.5.year or 8Syear followup. 1954-1962. Kahn (32).
a Study of 248.045 men aged 3184. at E-year or E-year followup. 1954-1969. Royot and Murray (49).
`Study of 51.911 men aged 18-69. at l&year followup. 1963-1972. Cederlof et al. (4).


People who persisted in cigarette smoking had more than twice the
risk of dying from CHD than those who quit even after taking into
account the other baseline differences. These studies provide stron-
ger evidence regarding the benefits of quitting than do the studies in
which all of the ex-smokers had stopped smoking before the
beginning of the followup.

Data from two "natural experiments" of smoking cessation among
physicians in Britain (12) and in California (14) are presented in
Table 3. Because these physicians have stopped smoking to a much
greater extent than has the general male population, the subsequent
CHD mortality trend in physicians as a whole relative to the general
population constitutes a crude estimate of the overall mortality
benefits of smoking cessation, This assumes that there have been no
other major risk factor changes in the compared populations, but
unfortunately, other risk factors were not measured in these two
studies. Both studies support the earlier prospective studies with
regard to the benefits of smoking cessation on CHD mortality. In
addition, they show the benefit of smoking cessation among a cohort
as a whole, including the continuing smokers with the quitters.
The most straightforward interpretation of ex-smoker data indi-
cating that CHD mortality rates of persons who stopped smoking are
substantially lower than those of persons who continued smoking, is
that smoking cessation directly results in the reduction of risk of
heart disease mortality. Underlying this presumed CHD benefit is
the assumption that ex-smokers are a representative sample of
smokers, except that they have stopped smoking. If the assumption
of representativeness is not valid and significant baseline differences
in relevant factors exist between ex-smokers and smokers, then the
mortality comparison of ex-smokers and continuing smokers may
not properly describe the benefits of smoking cessation for the
typical smoker. In the Kaiser-Permanente study (17), there were
small differences in risk profiles and other factors between those
who continued to smoke and those who quit, but these differences
were not large enough to account for the differences in CHD death
rates.

In summary, each of the several major prospective studies of
smoking cessation demonstrates that ex-cigarette smokers have a
decreased risk of subsequent mortality relative to continuing smok-
ers. The decreased risk occurs fairly quickly after cessation of
smoking, suggesting that the effects of cigarette smoking are
reversible. The quitters were self-selected in these observational
studies, however, and may include cigarette smokers at lower risk of
disease. However, the steadily decreasing risk over time after
quitting suggests that more is going on than the simple selection of a
lower risk group. Conversely, some smokers may quit in response to
symptoms or diagnosis of smoking-related illness, thus possibly

297


TABLE 2.-Age-, sex-, and race-adjusted death rates according to smoking category and selected major
          causes

Adjusted death rate per thousand person-years'

Category

No. of
subjects

No. of
person-
years

Ail causes

All causes
except mjuries
and poisoning

  All       Lung
neophms       cmm?r

  All
circulatory
diseasea

Persistent smokers

9,394      70,348       9.2 (557)        8.1 (465)        3.2 (191)      0.9 (58)       4.0 (240)      2.6 (166)

Tcamporary quitters        970       6,666       7.1  (46)         6.7  (43)        2.2  (14)       0.9  (6)       3.8  (24)       2.3  (16)

Persistent quitters       2,856      16,798       5.3 (107)         5.0 (102)        1.9  (39)       0.3  (6)       2.2  (46)       1.4  (31)

Never smokers        12,697      9%~       5.1 (569)         4.8 (540)        1.8 WW       0.02 (2)       2.4 (275)       1.6 W36)

' Figures m parentheses denote number of deaths.
Source: Fredmen et al. (17).


TABLE 3.-Relative trends in cigarette smoking and
     coronary heart disease mortality among male
      physicians in Britain and California in two
      natural experiments of smoking cessation,
       where status of other risk factors is unknown

British male physicians, 1951-71'

Percentage of physician current
smokers at start of time period
Ratio of smokers
(physicians/British males)

         Time period
1951-55    195660    1961-65    19671
-   -   -   -


  41        3.3        27        21


  88        68        al       51

Standardimd mortality ratio (physiciane/Britiah males)

CHD and myocard. degen., attained age
  Y20-54
  55-64
  65-74
  7&84
All causes, attained age
  2c-64
  65a4

107        85        62
120   -    103        a6
109       100        91
88        94       100


82        76        70
75        77        78

California male physicians, 1950-79'

Time period

                       KM-54 1955-59 1960-64 196.5-69 1970-74 1975-79
                       ------
Percentage of physician current
smokers at start of time period       53 48 39 28 20 14
Ratio of smokers
(phy&iaw/U.S. males)           100 a3 66 55 44 35

                    Standardized mortality ratio (phyeiciane/U.S. males)

Cl-ID                       115     97     86     80     74     69
Allcauaea                     a9     80     79     78     67     67

`Studyof31,CIOmenagsd20+.folfor~- DollandPeto (12)
`Study of 10,310 mea aged 26+, followed for 30 years. En&mm (14).

underestimating the benefits of quitting that would he expected in
an otherwise healthy population. Other variables that may contrib
ute to mortality may not have heen included in the analysis.

Randomized Controlled Trials of CHD Prevention Not Involving
Smoking Cessation

The most rigorous way to determine the value of smoking
cessation is the randomized controlled trial. A series of important
experimental or clinical trials have been conducted in the United
States and other countries over the past 25 years in order to

299


establish the effectiveness of primary prevention of CHD through
modification of risk factors. These randomized controlled trials
involve both primary and secondary prevention (2). The primary
prevention trials select subjects who are free of CHD or stroke at
entry to the study. The secondary prevention trials attempt to
modify risk factors after a heart attack or stroke in order to reduce
the risk of a second heart attack or death (6, 7, 8, 38, 40). Secondary
prevention trials and nonrandomized trials are not discussed further
here.

Most previous primary prevention trials of CHD have been limited
to a single risk factor such as serum cholesterol reduction. Many
single risk factor intervention trials include a pharmacologic agent
that lowers either serum cholesterol or blood pressure and is
compared with a placebo. Most of these studies are further limited to
higher risk subjects, such as subjects with serum cholesterol levels in
the highest 10 to 15 percent of the population, or to relatively small
sample sizes. They did not monitor or control for changes in cigarette
smoking habits.

The most extensive primary prevention trials involve dietary
reduction of cholesterol; they are described in more detail elsewhere
(2, 39). The major randomized trials are the Los Angeles veterans
domiciliary study (9), the Helsinki, Finland, mental hospital study
(42, 59), and a feasibility study of free-living and institutionalized
Americans (45). Each of these studies involved about 200 to 400 men
in the dietary intervention group and a similar number in the
control group.

Another set of randomized trials has involved reduction of high
blood pressure using antihypertensive medication-the U.S. Veter-
ans Administration cooperative study (631, the U.S. hypertension
detection and followup program (30, 31), the Australian therapeutic
trial (I), and the Oslo drug trial (24). These large studies followed
three small studies-Hamilton et al. (191, Wolff and Lindeman (67),
and the Cooperative Randomized Control Trial (CRCT) (5). These
studies generally show that lowered blood pressure results in some
reduction in CHD among the treated groups relative to the control
#FouP~

Intervention Trials of CHD Prevention Involving Smoking
Cessation

The observational epidemiological studies strongly suggest that
cigarette smoking cessation decreases the risk of heart attack and
CHD mortality compared with the risk for continuing smokers (60,
61, 62). All of the observational studies, however, have the limitation
that the individuals were not experimentally assigned to smoking
and nonsmoking status. Experimental studies such as randomized