HEALTH PROMOTION INTERNATIONAL
© Oxford University Press 1996
Vol. 11, No. 3
Printed in Great Britain
'No Smoking Class' competitions in Finland: their value
in delaying the onset of smoking in adolescence
ERKKI VARTIAINEN and ANNA SAUKKO
Department of Epidemiology and Health Promotion, National Public Health Institute, Helsinki,
Finland
MEM PAAVOLA
Youth Education Foundation, Helsinki, Finland
HARRI VERTIO
Cancer Society of Finland, Helsinki, Finland
SUMMARY
No Smoking Class' competitions have been organized
annually in Finland for eighth grade pupils since the
school year 1989-1990. Each school class has together to
decide if they want to be a no smoking class for the next
half a year from fall to spring. Classes decide about
participating in the contest and controlled non-smoking
during the competition; and classes where all the pupils
are able to refrain from smoking for the duration of the
competition participate in a lottery. The lottery prizes
consist of four main prizes, each of 10 000 FIM (US$
2000) and ten second prizes, each of 1000 FIM (US$
200). Each year, 20 000-30 000 pupils have participated
in the competition. The evaluation of the program was
carried out in the school year 1991-1992 with afollow-
up survey in 1993. For the evaluation, 97 classes were
randomly selectedfrom participating and non-participating classes. In 28 classes all the pupils were able to
remain non-smokers during the competition, 39 dropped
out from the competition because there were smokers in
the class and 30 classes were control classes, which did
not participate in the competition. Pupils filled in a
survey questionnaire on three occasions: before the
competition, and 1 month after and 1 year after the
competition. In the eighth grade, smoking increased
less in the classes which participated in the competition
than in classes which dropped out or were not participating in the competition. In the ninth grade, the onset rate
was similar in all groups of classes.
Key words: adolescence; prevention; smoking
INTRODUCTION
Smoking has not fallen significantly in the last
15 years in Finland (Rimpela et al., 1993). Moreover, smoking prevention programs have developed considerably in recent years (Glynn, 1989).
The most promising results have been obtained
from the social influence approach in school
health education (Flay, 1985; Bruvold, 1993).
Competition has been used in adults to help
them stop smoking (Korhonen et al., 1992) and
to introduce dietary changes to reduce blood
cholesterol levels (Vartiainen et al., 1992). The
aim of our study was to determine if competition
can be used to prevent the onset of smoking in
adolescence.
'NO SMOKING CLASS' COMPETITION
'No Smoking Class' competitions have been held
annually since the school year 1989-1990 for
eighth grade pupils (14-year-olds) in Finnish
schools. The competition material is sent to all
189
190 E. Vartiainen et al.
the 600 junior high schools in Finland. Classes
which want to participate in the competition
commit themselves to being a 'No Smoking
Class' for 6 months. Classes have themselves to
make the decision to participate. All the pupils
must agree to participate and not to smoke
during the program, and all the pupils sign a
commitment form promising not to smoke
during the competition. Classes monitor their
smoking by filling in a follow-up form every
week. At the end of the program, the forms of
the classes which remained non-smoking are sent
to the program office for the lottery. If someone
starts smoking during the competition and is not
able to quit, the class must drop out of the
competition. The responsibility of the control of
smoking rests mainly with the class itself. Classes
in which all pupils abstained from smoking until
the end of the program are allowed to participate
in a lottery. In the lottery there are four main
prizes, each of 10 000 FIM (US$ 2000) and ten
second prizes, each of 1000 FIM (US$ 200). The
prize money is given to the classes to use in any
appropriate way they choose. Often, classes have
used the money to finance an excursion or purchased equipment. There has been much attention paid in the national and local press to the
lottery and the uses to which classes have put the
prize money. Levels of participation in the competition can be seen in Table 1. Each class has a
contact teacher, who assists the class during the
competition and also organizes health education
sessions about smoking during school hours. The
program is organized by voluntary organizations
(Youth Education Union, Cancer Patients,
Cancer Society) and is financed by the Ministry
of Health and Social Affairs.
METHODS
The evaluation of the program was carried out
in the school year 1991-1992. In total, 1219
classes (23 012 pupils) entered the competition,
which is about one-third of the age cohort. Out
of these, 65 classes were randomly selected for
the study. The control group was selected randomly from the classes which did not enter the
competition. Pupils in these classes filled in
forms before the competition, and 1 month
and 1 year after the competition. This survey
was separate from the competition activities.
Questionnaires were anonymous; pupils placed
them in personal envelopes and the envelopes
Table 1: Levels of participation in the 'No Smoking
Class' competition in different school years
School year
Grade
Schools
Classes
Pupils
1989-1990
1990-1991
1991-1992
1992-1993
1993-1994
8th
8th
8th
8th
7th
8th
259
368
381
439
458
426
738
1219
1160
1329
1540
1181
14 765
23 012
20 653
26 084
30 747
23 325
were then sealed in a paper bag in the front of
the class and sent to the research team at the
National Public Health Institute. The pupils
were informed that no one would ever be able
to check their personal answers, and that the
envelopes would be opened by the study team
and not by their teachers or competition personnel. The pre-test was carried out before the
competition in fall 1991, the first post-test
1 month after the competition in spring 1992,
and the second post-test 1 year after the program in spring 1993. Participation in these three
tests is shown in Table 2. All the analyses utilized the following groups: those who participated in the competition to the end of the
program (28 classes in the sample); those who
dropped out from the competition (37 classes);
and the control group of those who did not
participate in the competition at all (30 classes).
The effect of the program was analyzed between
the group which originally decided to participate
in the competition (participants and drop-outs
pooled) and the control group. The analyses
were made in four ways: individual analyses of
those who had participated in all three surveys;
individual analyses including all the pupils in
each survey; and, using the class as an analysis
unit, using all the classes or using only the
classes which participated in all the surveys.
The conclusions were the same from the different types of analysis. Statistical analyses were
made by logistic regression analysis. In testing
the effects of competition on smoking, post-test
smoking status was the dependent variable and
participation in the competition and smoking
status before the program were independent
variables.
Smoking was assessed by the question: Do you
now smoke: not at all; less than once a month; 12 times a month; 1-2 times a week; daily. In the
analyses, those who smoked daily were regarded
as smokers.
'No Smoking Class' competitions in Finland 191
Table 2: Participation in the evaluation
Participation
Drop-outs
All
Controls
Pupils
Classes
Pupils
Classes
Pupils
Classes
Pupils
Classes
Pre-test
First post-test
Second post-test
580
549
557
28
28
28
704
613
677
37
33
37
551
531
519
29
28
28
1835
1693
1753
94
89
93
Participated in
all three surveys
489
28
487
31
443
27
1419
86
Table 3: Percentage of daily smokers in the pre-test, and in the first
and second post-test, and increase of smoking between the tests
Participants
Drop-outs
Participants
and drop-outs
Controls
Pre-test (a)
First post-test (b)
Second post-test (c)
n
5.2
7.5
16.1
489
10.8
14.7
21.2
487
8.0
11.1
18.8
976
11.3
16.4
22.5
443
Increase (%-points)
b-a
c-b
c-a
2.3
8.6
10.9
3.9
6.5
10.4
3.1
7.7
10.8
5.1
6.1
11.2
RESULTS
The onset of smoking was lower in those classes
which participated in the competition. Daily
smoking increased from the pre-test to 1 month
post-test by 2.3% percentage points, among
those pupils who participated in the competition,
3.9% percentage points in the group which
dropped out from the competition and 5.1%
percentage points in the control group. Among
those pupils who originally decided to participate
in the competition (participants and drop-outs),
smoking increased 3.1% percentage points.
Tested by logistic regression analyses, the odds
ratio between this group and control group was
1.55 (/? = 0.0268) using smoking at pre-test as a
covariate. This indicates that the onset of smoking was statistically significantly higher in the
control group. From the pre-test to the second
post-test, the increase in smoking was similar in
both groups. The odds ratio of 1.25 between the
groups was not statistically significant {p = 0.15).
Before the program, smoking was more
common in the control classes than in the classes
participating in the competition, and smoking
was almost as common in the classes which
dropped out during the program as in the control
classes (Table 3).
DISCUSSION
The 'No Smoking Class' competition has created
much publicity in the media. The program has
been relatively inexpensive, with an annual
budget of ~FIM 250 000 (~US$ 45 000). It
has reached about half of the birth cohort
(~ 30 000 pupils). The main weakness is the
large drop-out rate from the competition. Only
one-third of the classes have been able to remain
non-smokers for the duration of the competition.
Over the period evaluated, the best predictor of
drop-out was a high smoking rate at the beginning of the competition. The program is therefore
not feasible in classes where smoking is common
prior to the competition. Because of this, the
competition has been enlarged to include the
seventh grade, where regular smoking is less
common than in the eighth grade. At the beginning of the seventh grade, only a few pupils are
192 E. Vartiainen et al.
daily smokers, but the proportion increases to
~30% at the end of the ninth grade (Vartiainen
et al., 1990). The social pressure created by the
competition process was not greatly criticised in
pupils' answers. In the older age groups the
program can hardly be based on competition
between classes, but should be on an individual
or group basis. The program is also relatively
easy for schools and teachers to organize. In
Finland, health education is not an independent
subject in the curriculum, but is integrated with
other topics. This means that the infrastructure
for health education is relatively weak in Finnish
schools. Programs which involve a great deal of
work at schools are not usually feasible, especially if the program is not able to offer extra
resources for the schools, which is usually the
case in national health education programs.
The competition was innovative in two ways:
the control of the competition was in the hands of
the classes and one of the aims was to create a
continuous discussion on smoking among the
pupils. The decision-making process involved in
dropping out also allows discussion in class.
From the viewpoint of health promotion, these
aspects may prove to be as valuable as the issues
of not starting to smoke. To increase the validity
of self-reported smoking, the survey was separated from the competition. Pupils were not
aware that the evaluation survey was related to
the competition. All the questionnaires were
anonymous, and were put into envelopes by
each pupil and during the session were dropped
into a paper bag which was sealed by the pupils
and sent to the research team. The post-survey
was carried out 1 month after the competition
and not immediately. The aim of this was to
increase the validity of self-reporting of smoking.
The results of the post-survey show high levels of
honesty and self-control among the pupils, as
drop-out levels reflect the anonymously reported
levels of smoking.
The onset of smoking was less common in the
classes which decided to participate in the competition in the eighth grade, but in the ninth grade
the onset rates were similar in the competition
and control groups. As with many other public
health programs, scientists were not allowed to
decide who would participate in the program and
who would not. The only way to get any understanding of the possible effects of the program is
to use a quasi-experimental design as we have
used here. Because it was not possible to randomize classes to different conditions, we cannot be
sure that the observed effects were the result of
the competition. However, because the increase
was smallest in the eighth grade among those
who participated to the end of the competition
and similar to the ninth grade, the data suggest
that the competition itself delays smoking onset.
REFERENCES
Bruvold, W. (1993) A meta-analysis of adolescent smoking
prevention programs. American Journal of Public Health,
836, 872-880.
Flay, B. R. (1985) Psychosocial approaches to smoking prevention: a review of findings. Health Psychology, 4, 449488.
Glynn, T. (1989) Essential elements of school-based smoking
prevention programs. Journal of School Health, 59, 181188.
Korhonen, H. J., Niemensivu, H., Piha, T., Koskela, K.,
Wiio, J., Johnson, C. A. and Puska, P. (1992) National
TV smoking cessation program and contest in Finland.
Preventive Medicine, 21, 74-87.
Rimpela, A., Rimpela, M., Pohjanpaa, K., Karvonen, S. and
Siivola, M. (1993) Nuorten suojelu tupakkatuotteilla. Seurantaraportti 1977-1993. Nuorten terveystapatutkimus.
Kansanterveystiede, Oulun ja Helsingin yliopistot,
STAKES, Helsinki.
Vartiainen, E., Puska, P., Vlasoff, T., Muuttoranta, A.,
Korpelainen, V., Korhonen, H. J. and Nissinen, A. (1992)
Cholesterol lowering competition between small villages in
North Karelia (English summary). Sosiaalitaak Aikak, 29,
113-118.