Smoking and passive smoking in Estonia, Lithuania and Finland. Identifying target groups of tobacco policy. VILLE HELASOJA, RITVA PRATTALA, JURATE ...
Smoking and passive smoking in Estonia, Lithuania and Finland Identifying target groups of tobacco policy VILLE HELASOJA, RITVA PRATTALA, JURATE KLUMBIENE, JANINA PETKEVICIENE, ANU KASMEL, ANDRUS LIPAND, ANTT1 UUTELA, PEKKA PUSKA •
Keywords: Baltic, passive smoking, smoking, tobacco policy
T,
individual's own smoking or from environmental he prevalence of smoking in the Baltic countries is very tobacco smoke. Such risk groups related to poor behaviour similar to the situation in Finland during the 1960s. and living conditions are the most crucial targets for Smoking is thus common among males with an increase anti-tobacco action. This study compared Estonian, Finnish in the prevalence of smoking for females. One of the main and Lidiuanian prevalence rates and patterns of smoking features in the development of smoking in Finland from and passive smoking. To evaluate the possibilities of die 1960s onwards has been die reduction of daily smoking anti-tobacco policy, die concern of healdi effects of among men and die increase among women. ^ smoking was also assessed. The specific questions were as In populations widi substantial gender differences in follows. smoking prevalence passive smoking can be an important • What is die level of tobacco smoke exposure in diese part of exposure.^ Previous studies carried out in die Baltic countries? countries or in Eastern Europe have paid little attention • Is die level of tobacco smoke exposure related to gender, to passive smoking. As a consequence, a lack of informaage, education or urbanisation? tion exists on die total exposure to tobacco smoke. In Finland, exposure to passive tobacco smoke has decreased faster than the prevalence of daily smoking.5 One of die METHODS reasons for this is very likely die amendment of die This study was a part of FINBALT project, which has been tobacco legislation which came into force in 1995. This described in detail earlier.7 Data from cross-sectional legislation prohibits smoking in public places and workpopulation surveys of 1994 and 1996 from all diree sites.6 countries were used. The data were gadiered by mailed questionnaires which included questions on healdi status The aim of this study was to identify die population groups and on die use of healdi services, smoking, nutrition, use widi high exposure to tobacco smoke eitlier from die of alcohol, physical activity and traffic safety. The methodology and questionnaires of die surveys were care* V. HelMoja'. R. PrattalS . J. Klumbiene , J. Petkeviciene , A. Kasmel , fully standardised between die participating countries. A. Upand*. A. Uutrta , P. Puska The questionnaires were mailed in all countries in April1 National Public Health Institute, Department of Epidemiology and Health Promotion, Helsinki, Finland May and one or two follow-up reminders were sent. The 2 Institute for Bkxnedkal Research, Kaunas University of Medicine, Kaunas, response rates in 1994 were 83, 70 and 64% and in 1996 Lithuania diese rates were 77, 72 and 69% in Estonia, Finland and 3 Estonian Centre for Health Promotion, TaPmn, Estonia 4 Ministry of Social Affairs of Estonia, Tallinn, Estonia Lidiuania respectively. Comspondence: Ville Helasoja, National PuMk Hearth Institute, Gender, age, education and degree of urbanisation were Department of Epidemiology and Hearth Promotion, Mannerheimintie 166, chosen as die variables for characterising die population 00300 Helsinki. Finland, tel. +358 9 47441, f n +358 9 4744 8338, e-mail: vhehOktl.fi groups. The classification of diese variables and basic 1
1
2
1
2
3
Downloaded from eurpub.oxfordjournals.org by guest on May 24, 2011
Background: The aim of this study was to identify population groups which are the most crucial as targets for anti-tobacco action. Methods: A comparison was made of the prevalence and patterns of smoking and passive smoking in Estonia, Finland and Lithuania. Results: Total exposure to smoking, both one's own and passive smoking, was more common in two Baltic countries than in Finland. In these Baltic countries passive smoking was notably common among women. In all countries the exposure was more prevalent among the younger and less educated, but no difference emerged between urban and rural areas. Conclusions: The following target groups were identified as priorities for anti-tobacco actions: i) men of all ages in Estonia and Lithuania to stop their already established tobacco use, ii) young women in all three countries to prevent their starting a career of tobacco use and ill) young and less educated women in Estonia and Lithuania to prevent passive smoking, i.e. their exposure to environmental tobacco smoke.
Smoking and passive smoking
Table 1 Basic characteristics of the material (1994 and 1996) Men
Estonia Women
Total
Men
Finland Women
Total
34 (36) 33 (36) 33 (29)
32 (33) 38(39)
33 (33) 38 (39)
32 (33) 38(39)
29 (29)
1,736 873,980
30(27) 3,027 1561,986
Lithuania Women
Total
32(37)
34 (39)
34(34) 34(30) 2,151 1142,100
33 (34) 33 (27) 3,842 2198,000
43 33
23 36 42
23 39 38
1,702
2,174
3,876
42 23 35
46 23 31 2,162
44
Men
Age group (years) 20-34 (%) 35-49 (%) 50-64 (%) Total (n) Total ( n ) '
34 (38) 35 (36) 30 (26)
773
32 (34) 31 (36) 36(31) 963
418352
455,428
3383
29(28) 6,410
1533,268
3095,254
25 30
28 32
45
Education group (years) 0-9 (%)
22
19
20
31
10-12 (%)
48
42
45
30 766
39 953
35
34 35
1,719
2,986
3326
41 6312
Urbanisation group Cities (%) Towns (%)
39 22
42
40
25 36
23
39 762
40 21 39
39
Villages (%) Total (n)
41 21 39
35
951
1,713
3,027
3383
24 36 6,410
213 (%) Total (n)
36(41) 33 (34) 31 (25) 1,691 1055,900
23
1,695
23 33 3,857
a: The numbers in parenrbeses refer to the corresponding census data: Estonia as of 1 January 1996i, Finland as of 30 December 1996 and bthuanci as of 1 January 1997. Note that Estonian census data includes total Estonian population but material of rbis study consist of titular Estonians only.
Downloaded from eurpub.oxfordjournals.org by guest on May 24, 2011
characteristics of the material are presented in table 1. The iv) smoke-free persons (non-smokers not smoking age structure of the material was very similar for each passively). country when compared to the census data available; National data sets from Estonia, Finland and Lidiuania however, the oldest age group tended to be slightly overwere analysed separately widi SPSS. Basic characteristics represented. Education was measured as die total years were figured out by cross-tabulation widi y} tests. The of education from die questionnaire and was categorised socioeconomic patterning of tobacco smoke exposure was into die following diree groups: i) low = 9 years or less, compared by fitting logistic regression models to die ii) medium = 10-12 years and iii) high = 13 years or more. dichotomous- 'smoke-freeness' variable: i) 'smoke free' The chosen classification fits reasonably well into die (category iv of die typology) and ii) 'exposed' (categories educational systems of all die countries and yielded groups i—iii of die typology). All die models were fitted separately large enough for analysis. It should be noted diat, in diis to males and females in each country and die main effects study, 'Estonian' meant 'native (titular nationality) were included in their temporal order. The overall effect Estonian/Estonian speaking'. The results considering die was added first, followed by age, education and, finally, other nationalities of Estonia are commented on in die urbanisation. Models comprising the overall effect and text only if diey deviate significantly from titular each main effect only were also evaluated. The statistical Estonians. significance of the terms was assessed by die scaled deviance and die change of die degrees of freedom (ASD The variables describing die prevalences of smoking and and Adf). The results of die 'fully adjusted' models are passive smoking were 'smoking index' (daily smokers/ presented as odds ratios (ORs) and 95% confidence occasional smokers/quitters/never smokers), passive intervals. 11 smoking in home(somebody smokes/nobody smokes), passive smoking in workplace (over 1 h/less or none/not working outside home) and number of cigarettes smoked RESULTS daily (19 or less/20 or more). Concern for die negative The prevalence of daily smoking was highest in Estonia. healdi effects of smoking was analysed widi smokers' Occasional smoking was between 6 and 8% in all counconcern of die detrimental effects of die smoking (yes/no) tries. The prevalence of quitting was more substantial in and smokers' desire to quit smoking (yes/no or not sure). Finland (20%) than in Estonia (10%) or Lithuania (8%). Daily smoking (daily smokers/odiers) and passive The prevalence of never-smoking was highest in smoking (in home or over 1 h in workplace/not at all) Lithuania. In addition, die proportion of daily smokers were combined together as a typology: i) exposed smokers who smoked over one pack a day was lowest in Lithuania. (daily smokers smoking also passively), ii) non-exposed The differences in the prevalence of daily smoking smokers (daily smokers not smoking passively), iii) exbetween the sexes were largest in Lithuania and smallest posed non-smokers (non-smokers smoking passively) and in Finland. Among Lidiuanian females, die prevalence of
EUROPEAN JOURNAL OF PUBLIC HEALTH VOL 11 2001 NO. 2
Men
100% 80%
Estonia
Finland
Lithuania
• Exposed smokers (daily smokers smoking also passively) D Non-exposed smokers (daily smokers not smoking passively) • Exposed non-smokers (non-smokers smoking passively) Q Smokefree persons (non-smokers not smoking passively)
Women
100% 80% 60% 40% 20% 0%
Figure 1 Types of tobacco smoke exposure in 1994 and 1996
educated in all countries. The differences between urban and rural areas were small and inconsistent. In addition, the pattern was similar among men and women. In all countries daily smokers were equally concerned with the detrimental effects of smoking for their health (Estonia and Finland both 77% and Lithuania 75%). Smokers likewise also expressed their desire to give up smoking very similarly (Estonia 59%, Finland 56% and Lithuania 69%). The age, educational and regional differences in those who expressed concern for their own smoking and the willingness to give up were in general small and no uniform pattern emerged.
Table 2 Prevalence of daily and passive smoking in 1994 and 1996 Men
Estonia Women Total
Daily smokers (%) Total (n)
52 760
21
35
952
1,712
Passive a: home (%) Total (n)
52 759
45 948
1,707
47 674
26 758
1,432
Passive at work (%) b Total (n)
(95% CI)
48
35
a: Lithuanian passive smoking at home in 1996 only \x Working outside home only
(33-37)
(46-50)
(33-37)
Men
Finland Women Total
30
19
24
2,931
3318
6,249
27
20
24
2,991
3342
6333
23
10
16
2386
2,465
4,851
(95% CI) (23-25)
(23-25)
(15-17)
Men
Lithuania Women Total
47
8
25
1,621
2,084
3,705
46 893
42
44
1,063
1,956'
38
16
27
1,426
1,590
3,016
(95% CI) (24-26)
(42-46)
(25-29)
Downloaded from eurpub.oxfordjournals.org by guest on May 24, 2011
never smoking was higher (83%) than in Estonia or Finland (64 and 61% respectively). Age and educational differences in daily smoking were broadly similar in all countries, with the prevalence being highest in the youngest and lowest in the oldest groups. Furthermore, there were less daily smokers in the 'highest' education groups. Among females, the oldest age groups were most often never smokers (Estonia 80%, Finland 76% and Lithuania 95%). Passive smoking at home was more common in Estonia and Lithuania than in Finland. In all countries men, younger and the less educated lived more often in an environment where somebody smoked. The proportion of persons smoking passively at their workplace was also higher in Estonia and Lithuania than in Finland (table 2). In Estonia and Lithuania the proportions of totally smoke-free persons (40 and 44% respectively) and exposed non-smokers (25 and 29% respectively) were very similar. In Lithuania, 38% of the women were exposed non-smokers. On the contrary, in Finland the proportion of smoke-free persons was 65% and the proportion of exposed non-smokers was 11 % (figure I). Logistic regression models confirmed that the distribution of 'smoke freeness' was similar in all countries. This was mainly connected with age and education, but not with urbanisation. Adjustment had only a marginal impact on the results. Because the same conclusions could be drawn from the unadjusted and fully adjusted models, only the results of the fully adjusted models are shown in table 3. In the unadjusted models comprising the overall effect plus each explanation, the effects of age and education were statistically significant (p
