Andreeva, T. I., & Krasovsky, K. S. (2011). Recall of tobacco pack health warnings by the population in Ukraine and its association with the perceived tobacco health hazard. Int J Public Health, 56(3), 253-262.
Recall of tobacco pack health warnings by the population in Ukraine and its association with the perceived tobacco health hazard
Tatiana I Andreeva 1, Konstantin S Krasovsky 2
1School of Public Health, National University of Kyiv-Mohyla Academy, Kiev, Ukraine
2Department of tobacco control, Institute for strategic research, Ministry of Health of Ukraine, Kiev, Ukraine
Abstract
Objectives
After years of prevalence of smoking increase, Ukraine observes its decline. Recent tobacco control measures included smoke-free policies, new textual health warnings (THW) since late 2006, ban of outdoor tobacco advertising since January 2009 and tobacco tax increase since late 2008. The objective was to estimate potential
contribution of THW to smoking decline process in Ukraine.
Methods
The study is based on a nationwide omnibus survey of 2008 Ukrainian adults (18+).
Outcome measures: quitting smoking after 2006 and perception of tobacco-related
hazards. To measure the exposure to THW respondents were asked to describe the warnings they recall. Multivariate regression analysis was performed in SPSS 15.0.
Results
Those who considered health hazard of smoking as serious were significantly more likely to quit. Male smokers were more likely to perceive health hazard if they recalled health warnings 'Smoking is addictive, do not start to smoke!' and 'Smokers die early'.
Conclusions
THW reach those groups of the population they are aimed to and may potentially result in consequent quitting smoking in male smokers.
Key words: tobacco smoking; smoking cessation; tobacco control; cigarette packages health warnings.
Text
Introduction
The Framework Convention on Tobacco Control (FCTC) recommends evidence-based measures to reduce tobacco use. However, their practical implementation in countries with high smoking prevalence can benefit from particular information on how every single measure reaches different socio-demographic groups, and what can be done to increase the effect.
Before Ukraine enacted the first tobacco control law in late 2005 and ratified the FCTC in 2006, prevalence of smoking was increasing. The age-standardised prevalence of current smoking in Ukrainian men was 54.8% in 2001, and 66.8% in 2005. In Ukrainian women, prevalence increased from 11.5% in 2001 to 20.0% in 2005. (Andreeva and Krasovsky 2007) However, after the abovementioned legislative acts Ukraine observes decline in smoking prevalence. Between June 2005 and May 2009 prevalence of daily smoking among adults changed from 37% to 27%, in men from 62% to 49%, in women from 17% to 9%, as was seen in several nationwide omnibus surveys (Krasovsky et al.
2009; Andreeva et al. 2009) and recently confirmed by the Global Adult Tobacco Survey data.(Ukraine Global Adult Tobacco Survey report 2010) The smoking prevalence decline is seen in most socio-demographic groups, and in all regions. The decline in women was already significant in 2006 compared to 2005, while in men only slight decline was revealed in 2007, which makes us hypothesize that smoking men and women responded differently to the undertaken measures. Since 2005, several tobacco control policies which could contribute to the decline of smoking prevalence were implemented in Ukraine. Smoke-free policies (Goodman et al. 2009) supported by
media campaign since the middle of 2006 covered many workplaces and public places.
At least 50% of the area of restaurants and bars had to be isolated from the smoking area so that the tobacco smoke does not penetrate to the smoke-free areas. New textual health warnings labels (NTHWL) covering 30% of the surface of two larger sides of cigarette packs (one main warning on the front side related to cancer and cardiovascular diseases and one of six additional warnings on the back side related to impotence, addiction, smoking in pregnancy, second-hand smoke impact in kids and adults, and premature deaths of smokers) since late 2006 substituted a warning born by all packs and covering just 10% of the front surface ‘Ministry of Health warns: smoking is bad for your health’ without specific requirements regarding the colours of the text warning.
Smoking cessation services implemented in some territories of Ukraine could also contribute a bit. Ban of outdoor tobacco advertising since January 2009 could also somewhat influence at later stages of smoking prevalence decline under consideration along with tobacco tax increase since late 2008 which accounted for 50% price increase by the time of the survey compared to one year earlier.
While it is well-known that all the tobacco control measures potentiate each other, still it could be helpful in achieving ultimate results to assess how different demographic groups react to particular tobacco control measures.
The aim of this study is to estimate whether there was a significant contribution of NTHWL introduced since late 2006 in the decline of smoking prevalence among the Ukrainian population, and if there was, which particular groups responded most. FCTC Article 11 requires each Party to adopt and implement, within a period of three years after entry into force of the FCTC for that Party, effective measures to ensure that tobacco product packaging and labelling carry large, rotating health warnings (which
must describe the harmful effects of tobacco use, and may include other appropriate messages) that should cover at least 50% – and must cover at least 30% – on the principal display areas (usually the front and back of the pack) (Article 11.1(b)).
Health warnings on tobacco product packages are known to be effective in increasing the perception of health risk, supporting the intention to quit tobacco use, discouraging the intention to begin tobacco use, and increasing cessation rates (CDC 2009;
Hammond et al. 2003). Health warnings make cigarette packs less attractive, distract the attention from the particular brand and increase the perceived risk of dying from
smoking-related illness, especially to smokers who already intended to stop smoking (Willemsen 2005; Miller et al. 2009). Still, not all the cigarette package warning labels are effective in building awareness regarding tobacco health impact. For example, the health warnings in the USA were claimed to be ineffective (Krugman et al. 1999; Givel 2007), or less effective than could be (Nimbarte et al. 2005).
Most of the studies which show positive effects of health warnings were considering graphic or pictorial ones (Borland et al. 2009a; Etter and Cornuz 2009; Hammond et al.
2004a; Hammond et al. 2003; Hammond et al. 2004b; Thrasher et al. 2006; Thrasher et al. 2007a; Thrasher et al. 2007b; White et al. 2008). With regard to textual health warnings, tobacco consumption decline was linked to the introduction of such warnings in Turkey, South Africa, Poland and Canada (Curbing the epidemic: Governments and the economics of tobacco control 1999); however, detailed studies related to textual health warnings were less numerous, and were conducted in Australia (Borland 1997) and the European Union (Willemsen 2005). It was also shown that the graphic warnings may be superior to text-based warnings as pictorial health warnings stimulated more cognitive responses (Borland et al. 2009a; CDC 2009; Hammond et al. 2007; Hammond
et al. 2006; O'Hegarty et al. 2006; O'Hegarty et al. 2007; Peters et al. 2007) with most of these studies comparing either Canadian graphic warnings to US textual ones or Australian warnings to the UK text-only warnings. In one study (Hassan et al. 2008) textual health warnings in the UK and the US were compared showing that more prominent textual warnings implemented by the European Union resulted in more contemplation of quitting in smokers. As all these studies were conducted in Western, high-income countries, there is little evidence for such an impact in other countries.
This paper addresses an important research question of the consistency of textual warnings effects across cultural, political and socio-demographic settings.
We aimed to measure how much the health warnings introduced in Ukraine are seen by the population, and whether recall of health warnings is associated with tobacco health impact awareness or intention and attempts to stop smoking. We hypothesized that certain socio-demographic groups are better reached by health warnings than the others, that remembering particular or all health warnings is associated with perception of higher tobacco-related harm, and that such perception is associated with attempts to stop smoking.
Methods
The study is based on a nationwide omnibus survey of 2008 randomly selected Ukrainian adults (18+) between May 21 and May 31, 2009, conducted by the Kyiv International Institute of Sociology (KIIS). Interviews were conducted in person in Ukrainian or Russian according to respondents’ preference. KIIS omnibus surveys are nationwide regular surveys of adult population of Ukraine aged 18 years and over with
four-stage sampling scheme and random sampling at each stage. The survey was
conducted in 110 settlements of all 26 administrative territories of Ukraine based on the electoral lists. The sample accurately reflected the demographic and geographic profile of the country. The margin of error for the sample of 2000 respondents is +/- 2.2% at the 95% level of confidence.
As an ultimate outcome measure we considered quitting smoking in particular years, and the perception of tobacco-related hazards was considered as an intermediary outcome.
Smoking status was measured according to the WHO ‘Monitoring the tobacco
epidemic’ 1998 publication (WHO 1998) with identifying seven categories of smoking status: daily smokers, occasional smokers, reducers, former daily smokers, former occasional smokers, experimenters, never smokers. For particular purposes, seven groups of smoking status were collapsed into three – current smokers (daily, occasional and reducers), former smokers (former daily and former occasional smokers) and non- smokers (experimenters and never smokers).
To assess the impact of particular measures, ex-smokers were asked about the year when they quit for distinguishing those who quit smoking after the introduction of health warnings in late 2006. For these purposes, in a nested case-control design those who reported to quit smoking in 2007, 2008 and 2009 were considered cases and those who were still current smokers in May 2009 were considered controls.
Perception of tobacco-related hazards was measured with a question ‘Now, thinking about the possible health effects of smoking cigarettes, in general, do you think that smoking cigarettes is a serious health hazard, a moderate health hazard, a minor health hazard, or not really a health hazard at all?’ As the options b-e (b. Moderate health
hazard, c. Minor health hazard, d. Not a health hazard at all, e. Don’t know/undecided) were less numerous than the first answer to this question (a. Serious health hazard) and were rather homogeneous in terms of their association with the intent and the attempts to quit, the options b-e were collapsed into one alternative.
To measure the respondent’s attention to health warnings, study participants were asked to describe the warnings they recall. Interviewers were instructed not to show the texts to the respondents and not to expect the exact wording of the warning, but to mark the particular warning as ‘recalled’ given the respondent mentions the corresponding topic (for instance ‘impotence’, ‘heart disease’, ‘pregnancy’ etc.). Number of the warnings that were recalled was also used as a measure of the respondent’s attention to health warnings with minimum number of health warnings recalled equal to 0, and maximum equal to 7.
To control for the potential impact of socio-demographic factors, information on gender, age, and education was collected as well. Education was described as one of four options: incomplete secondary or less, complete secondary, vocational secondary, and complete higher.
Regression of the number of recalled health warnings on socio-demographic variables was analysed with the use of the General Linear Model option within SPSS 15.0. Main effects and interaction of separate categorical variables were analysed.
For identifying the factors related to the recall of separate warnings, multivariate binary logistic regression analysis was performed separately for current smokers, former smokers and non-smokers.
Associations of recent quitting smoking and perception of tobacco health hazard with other categorical variables were analyzed with the use of multivariate binary logistic regression analysis as well.
Results
Percentage distribution of the study sample is shown in Table 1. Females constituted 59% of the sample, 54% were never smokers and 25% were daily smokers.
Who notices and recalls health warnings? Sum of warnings
In all demographic groups, minimum number of NTHW recalled was 0, and maximum equalled 7, while the average was 2,24. Multivariate comparison of means between different groups of respondents with the use of General linear model is shown in Table 2.
Smokers on average recalled more warning labels than non-smokers. More educated people recalled more health warnings than those with lower education. However, an interaction with gender was found – in men those less educated (who are more likely to smoke) recalled more NTHWL.
Recall declined with age from 3.05 in 25-34 years old group to 1.48 in those aged over 55. However, the decline by age was only typical for non-smokers and former smokers, while the decline was smaller in current smokers.
Particular warnings
Besides the number of NTHWL recalled, probability of remembering every particular health warning was analyzed as well. Percentages of respondents who recalled the
warnings are shown in Table 3, and this was done for the whole sample as well as for current smokers, former smokers, and non-smokers separately.
Those warnings which are placed on the front side of the every pack ('Smoking causes cardiovascular diseases and lung cancer') were recalled most frequently – by 70% of respondents. Among other six warnings two warnings were recalled by about one third of respondents ('Smokers die early' and 'Tobacco smoke harms the health of those around you'), two other health warnings were recalled by one in four respondents ('Smoking when pregnant harms your baby' and 'Smoking causes impotence') and two warnings by less than one in five respondents ('Protect children: do not make them inhale your smoke' and 'Smoking is addictive, do not start to smoke!')
Smokers were more likely to recall every warning than former smokers and non- smokers. Higher odds for smokers to recall a health warning compared to non-smokers were more prominent (OR 3,37 (95% CI 2,56-4,42)) for the front-side warning and the weakest for the warnings related to passive smoking ('Protect children: do not make them inhale your smoke' – OR 1,23 (95% CI 0,96-1,57), and 'Tobacco smoke harms the health of those around you' – OR 1,35 (95% CI 1,09-1,67)).
While some health warnings were equally recalled by males and females, warning related to consequences of smoking in pregnancy was more likely to be recalled by women-smokers than men-smokers OR=1,35(1,06-1,72), and other three warnings ('Smoking is addictive, do not start to smoke!', 'Smoking causes impotence', and 'Smokers die early') were recalled by larger proportion of men than women who do not smoke.
In all smoking status groups, younger respondents recalled health warnings more likely than those older ones. However, there is no such association in current smokers with
regard to harm of SHS, thus older smokers are more likely to pay attention to the information related to SHS than the younger ones.
Association between recall of health warnings and education is seen only in non-
smokers. Thus smokers with different education are equally reached by the information in health warnings.
Association between health warnings-recall, tobacco harm perception, and quitting smoking
Only 38 respondents said they quit smoking since the beginning of 2007. They were compared to those 630 who were still current smokers by May 2009. No significant association with remembering particular health warnings was found. However, among those who considered tobacco-related health hazard as serious 31 (7.0%) quit smoking in 2007-2009, while among the others only 7 (2.5%) with OR= 3,362 (95%CI 1,458- 7,750). The association was significant only for male smokers.
Associations of particular warnings recall with perception of health hazard of smoking were explored in men stratified by smoking status and results of bivariate analysis are presented in Table 4. Smokers were more likely to perceive health hazard of tobacco use as serious if they recalled health warnings 'Smoking when pregnant harms your baby', 'Smoking is addictive, do not start to smoke!', 'Protect children: do not make them breathe your smoke', and 'Smokers die early'. In multivariate analysis recalling NTHWL related to addictiveness of cigarettes and early deaths of smokers were independently associated with the perception of tobacco health hazard. None of the health warnings were associated with higher perception of tobacco-related harm in former smokers, and
recalling the warning ‘Smokers die early’ was associated with higher perception of harm in non-smokers.
Discussion
The study assessed how the textual messages introduced in Ukraine in line with the FCTC requirements are seen by the population and whether noticing these messages was associated with higher awareness of tobacco-related harm.
As front-side health warnings are being exhibited at the points of sales and in advertising we could expect them to be seen and recalled more than other health warnings by both smokers and non-smokers. However, this warning ('Smoking causes cardiovascular diseases and lung cancer') is definitely aimed at smokers, and we found its recall to be stronger associated with the smoking status than the recall of other warnings. On the other hand, passive smoking-related warnings are more evenly recalled by both smokers and non-smokers. This means that the intended groups are reached by those warning labels. Besides, warnings related to pregnancy are more likely recalled by women, while warnings about impotence are more recalled by men. This gives information about the selective attention of different segments of the population to those NTHWL which were aimed at them. Thus, we conclude that by means of
measuring the recall of health warnings we get information not just about presence but about the potential impact of health warnings.
High recall of health warnings among young non-smokers which was almost equal to the recall of daily smokers could reflect the attention of young people to tobacco packs health warnings and their potential role in preventing smoking initiation among the
youth. Some studies suggest that health warnings may have the potential to lower smoking intentions of adolescents (White et al. 2008). Large portion of recent smoking prevalence decrease was probably due to lower initiation; however, the role of NTHWL in this requires additional research.
Males were found to react more to health warnings than females. For male-smokers it was found that recalling several warnings was associated with higher perceived health hazard related to tobacco, which could be a trigger towards quitting smoking. No association found in female smokers could be partly due to their smaller number in the sample.
Indirect evidence which contributes to the understanding that health warnings in
Ukraine worked more for quitting smoking in male smokers is the following fact. Some slight decline in smoking prevalence in males started to be seen only in 2007 compared to 2006, while significant decline of prevalence in females was already found in 2006 compared to 2005 (Andreeva et al. 2009). This suggests that female smokers responded more to the smoke-free policies-related information campaign, while male smokers reacted more after the introduction of new health warnings. Thus, we see that some of the health warnings could contribute to higher awareness of tobacco-related harm in male smokers and possibly to their quitting attempts. While studies of the effectiveness of health warnings are not numerous, International Tobacco Control Policy Evaluation Study (ITC) has shown that noticing health warnings was associated with higher knowledge about tobacco risks among smokers (Hammond et al. 2006). Though the direct association between NTHWL recall and quit attempts was not found in our study, the results suggest that an indirect link may be present. On the one hand we revealed a link between recall of health warnings and perception of tobacco related hazard. On the
other hand, this perception was associated with quitting. Obviously, noticing a warning does not lead to immediate quitting; however, it may predict downstream quit attempts.
Other studies have shown the so-called ‘prospective predictors of making quit attempts’
(Borland et al. 2009a; Borland et al. 2009b).
Our study found which exactly health warnings were associated with perception of health hazard in male smokers. These were warnings regarding addiction ('Smoking is addictive, do not start to smoke!') and premature death ('Smokers die early') as well as two warnings related to the harm to children through second-hand smoke exposure and smoking in pregnancy. Probably this was exactly the truth which was not obvious to smokers earlier. This is likely to illustrate the idea that the goal of health warnings is the continuous dissemination of the latest scientific findings on the health consequences of smoking (Potschke-Langer and Schulze 2005). The effects of the health warning regarding addiction are expected based on what is known from the international experience. In Australia tobacco industry paid special efforts to counteract the introduction of this specific health warning (Chapman and Carter 2003). However linking higher awareness to exclusively health warnings has its limitations, as simultaneously there were other information efforts to educate the smokers.
In our study we assessed effects of textual black and white health warnings, not graphic ones mostly evaluated in recent studies related to health warnings effectiveness. Though textual warnings are known to be less effective than the pictorials, Borland R. has shown effects of textual warnings introduced in Australia in 1995 as a means of
stimulating thoughts about negative consequences of smoking, refraining from smoking a planned cigarette, and making a quit attempt (Borland 1997). In 1998, textual 30%
health warnings were introduced in Poland resulting in 14% of smoking men and 16%
of smoking women attempting to cut down smoking, and 3% of men and 4% of women said they had stopped smoking under the influence of such warnings (WHO 2009).
Comparison of the effects of textual health warnings in the UK and the USA (Hassan et al. 2008) has shown that more prominent textual health warnings introduced by the European Union were more likely to lead to contemplation of quitting in smokers.
Similar type of warnings was under our consideration.
Whether health warnings implemented in Ukraine in 2007 will have long-lasting effect is not known. Studies conducted in the ITC Policy Evaluation Project countries have shown that changes in health warnings are associated with effects (Hammond et al.
2007), which can probably decrease further.
On average, health warnings were found to be more likely recalled by younger and more educated people. This may be due to the fact that younger people have better memory and people with more formal education had opportunity to train their
intellectual functions more than those who did not graduate colleges and universities.
However, this difference is more expressed in non-smokers, which means that tobacco packs health warnings reach those who would be difficult to educate with other health promotion measures.
The strength of the study was that the participants were asked to recall particular health warnings which gave the information not found in other studies in the area known to the authors.
The study bears all the limitations of the cross-sectional design. Most of the favourable consequences of health warnings cited above were found in follow-up studies. In our survey a reverse order of events (‘Have you quit in 2007?’-‘Do you recall health warnings in 2009?’) poses even more limitations. Expectably, those who still smoke
recalled more health warnings than those who successfully quit a year or two ago, which does not allow establishing the association.
While the whole sample was nationally representative, subsample of smokers and especially female smokers could be insufficient for drawing conclusions with enough statistical power.
Measuring recall of health warnings could vary between interviewers. This could be a substantial cause of uncertainty in our study. Besides, some of the questions typically asked in the surveys related to health warnings effects like about postponing a planned cigarette were not asked in the survey due to logistics reasons.
With all the limitations listed above as well as the absence of the comparison group and the pre-policy assessment, the study does not give grounds for establishing causal relationship between NTHW and smoking prevalence decline, but just suggests hypotheses for further consideration.
Most of the limitations of this study could be overcome in a cohort study aimed to document responses of smokers to different tobacco control measures. However, limited resources make researchers in low and middle income countries rely on cheaper study designs.
Conclusions
1. More prominent multiple textual health warnings introduced in Ukraine since late 2006 have likely contributed to building tobacco-and-health awareness among Ukrainian population.
2. Noticing and recall of health warnings reflects their potential impact on the population awareness and behaviour: passive smoking related warnings reach non- smokers as well, pregnancy-related warnings reach women more than men, impotence-related warnings reach more men than women, etc.
3. Health warnings-remembering was stronger associated with the perception of health hazard of tobacco use among male smokers than females, and could potentially result in consequent quitting smoking.
4. Several warnings could be considered as more effective with regard to making male smokers think about quitting: 'Smoking when pregnant harms your baby', 'Smoking is addictive, do not start to smoke!', 'Protect children: do not make them breathe your smoke', and 'Smokers die early'.
5. Health warnings could contribute to the deterring young people from starting smoking or to stopping in the experimentation phase before daily smoking is established.
Recommendations
The study shows potential contribution of cigarette packages health warnings in prevention smoking initiation and promoting smoking cessation. Though with these health warnings Ukraine meets the requirements of the FCTC, experience of other countries show that cigarette packages whose warning labels contain prominent graphic imagery are more likely than text-only warning labels to promote smoking-related knowledge and smoking cessation (Thrasher et al. 2007a). International experience shows that the dissemination of Quitline phone numbers in combination with health warnings on every packet clearly improves the effectiveness and broad reach of smoking cessation advice via Quitlines (Potschke-Langer and Schulze 2005).
Acknowledgements
The data collection was performed by the Kyiv International Institute of Sociology and supported with funds from the Campaign for Tobacco Free Kids within the Bloomberg Global Initiative to reduce tobacco use.
Conflict of interest
The authors declare that they have no competing interests.
References
Andreeva TI, Krasovsky KS (2007) Changes in smoking prevalence in Ukraine in 2001- 5. Tob Control 16 (3):202-206. doi:16/3/202 [pii]
10.1136/tc.2006.019588 [doi]
Andreeva TI, Krasovsky KS, Kharchenko NM (2009) Correlates and recent changes of smoking prevalence among adults in Ukraine (in Ukrainian). Eastern European Journal of Public Health (1):50-57
Borland R (1997) Tobacco health warnings and smoking-related cognitions and behaviours. Addiction 92 (11):1427-1435
Borland R, Wilson N, Fong GT, Hammond D, Cummings KM, Yong HH, Hosking W, Hastings G, Thrasher J, McNeill A (2009a) Impact of graphic and text warnings on cigarette packs: findings from four countries over five years. Tob Control 18 (5):358-364. doi:tc.2008.028043 [pii]
10.1136/tc.2008.028043
Borland R, Yong HH, Wilson N, Fong GT, Hammond D, Cummings KM, Hosking W, McNeill A (2009b) How reactions to cigarette packet health warnings influence quitting: findings from the ITC Four-Country survey. Addiction 104 (4):669- 675. doi:ADD2508 [pii]
10.1111/j.1360-0443.2009.02508.x [doi]
CDC (2009) Health warnings on tobacco products - worldwide, 2007. MMWR Morb Mortal Wkly Rep 58 (19):528-529. doi:mm5819a3 [pii]
Chapman S, Carter SM (2003) "Avoid health warnings on all tobacco products for just as long as we can": a history of Australian tobacco industry efforts to avoid, delay and dilute health warnings on cigarettes. Tob Control 12 Suppl 3:iii13-22 Curbing the epidemic: Governments and the economics of tobacco control (1999).
World Bank Publications. World Bank,
Etter JF, Cornuz J (2009) [Are graphic warnings on cigarette packs useful?]. Rev Med Suisse 5 (210):1476-1479
Givel M (2007) A comparison of the impact of U.S. and Canadian cigarette pack warning label requirements on tobacco industry profitability and the public health. Health Policy 83 (2-3):343-352. doi:S0168-8510(06)00290-9 [pii]
10.1016/j.healthpol.2006.12.003 [doi]
Goodman PG, Haw S, Kabir Z, Clancy L (2009) Are there health benefits associated with comprehensive smoke-free laws. Int J Public Health 54 (6):367-378.
doi:10.1007/s00038-009-0089-8 [doi]
Hammond D, Fong GT, Borland R, Cummings KM, McNeill A, Driezen P (2007) Text and graphic warnings on cigarette packages: findings from the international tobacco control four country study. Am J Prev Med 32 (3):202-209. doi:S0749- 3797(06)00529-0 [pii]
10.1016/j.amepre.2006.11.011 [doi]
Hammond D, Fong GT, McDonald PW, Brown KS, Cameron R (2004a) Graphic Canadian cigarette warning labels and adverse outcomes: evidence from Canadian smokers. Am J Public Health 94 (8):1442-1445. doi:94/8/1442 [pii]
Hammond D, Fong GT, McDonald PW, Cameron R, Brown KS (2003) Impact of the graphic Canadian warning labels on adult smoking behaviour. Tob Control 12 (4):391-395
Hammond D, Fong GT, McNeill A, Borland R, Cummings KM (2006) Effectiveness of cigarette warning labels in informing smokers about the risks of smoking:
findings from the International Tobacco Control (ITC) Four Country Survey.
Tob Control 15 Suppl 3:iii19-25. doi:15/suppl_3/iii19 [pii]
10.1136/tc.2005.012294 [doi]
Hammond D, McDonald PW, Fong GT, Brown KS, Cameron R (2004b) The impact of cigarette warning labels and smoke-free bylaws on smoking cessation: evidence from former smokers. Can J Public Health 95 (3):201-204
Hassan LM, Shiu E, Thrasher JF, Fong GT, Hastings G (2008) Exploring the
effectiveness of cigarette warning labels: findings from the United States and United Kingdom arms of the International Tobacco Control (ITC) Four Country Survey. International Journal of Nonprofit and Voluntary Sector Marketing 13 (3):263-274
Krasovsky KS, Andreeva TI, Grigorenko AA, Butylska NA (2009) Assessment of dynamics in tobacco smoking prevalence in Ukraine (in Ukrainian). Bullitin of Social Hygiene and Health Care in Ukraine (2):91-99
Krugman DM, Fox RJ, Fischer PM (1999) Do cigarette warnings warn? Understanding what it will take to develop more effective warnings. J Health Commun 4 (2):95-104
Miller CL, Hill DJ, Quester PG, Hiller JE (2009) Response of mass media, tobacco industry and smokers to the introduction of graphic cigarette pack warnings in Australia. Eur J Public Health. doi:ckp089 [pii]
10.1093/eurpub/ckp089 [doi]
Nimbarte A, Aghazadeh F, Harvey C (2005) Comparison of current U.S. and Canadian cigarette pack warnings. Int Q Community Health Educ 24 (1):3-27.
doi:9PX0NBG10ALAG5YH [pii]
10.2190/9PX0-NBG1-0ALA-G5YH [doi]
O'Hegarty M, Pederson LL, Nelson DE, Mowery P, Gable JM, Wortley P (2006)
Reactions of young adult smokers to warning labels on cigarette packages. Am J Prev Med 30 (6):467-473. doi:S0749-3797(06)00110-3 [pii]
10.1016/j.amepre.2006.01.018 [doi]
O'Hegarty M, Pederson LL, Yenokyan G, Nelson D, Wortley P (2007) Young adults' perceptions of cigarette warning labels in the United States and Canada. Prev Chronic Dis 4 (2):A27. doi:A27 [pii]
Peters E, Romer D, Slovic P, Jamieson KH, Wharfield L, Mertz CK, Carpenter SM (2007) The impact and acceptability of Canadian-style cigarette warning labels among U.S. smokers and nonsmokers. Nicotine Tob Res 9 (4):473-481.
doi:776190417 [pii]
10.1080/14622200701239639 [doi]
Potschke-Langer M, Schulze A (2005) [Health warnings on cigarette packets. An overview]. Bundesgesundheitsblatt Gesundheitsforschung Gesundheitsschutz 48 (4):464-468. doi:10.1007/s00103-005-1020-y [doi]
Thrasher JF, Allen B, Reynales-Shigematsu LM, Anaya R, Lazcano-Ponce E, Hernandez-Avila M (2006) [Analysis of the impact of cigarette pack graphic
warnings on Mexican smokers]. Salud Publica Mex 48 Suppl 1:S65-74.
doi:S0036-36342006000700008 [pii]
Thrasher JF, Hammond D, Fong GT, Arillo-Santillan E (2007a) Smokers' reactions to cigarette package warnings with graphic imagery and with only text: a
comparison between Mexico and Canada. Salud Publica Mex 49 Suppl 2:S233- 240. doi:S0036-36342007000800013 [pii]
Thrasher JF, Rousu MC, Anaya-Ocampo R, Reynales-Shigematsu LM, Arillo-Santillan E, Hernandez-Avila M (2007b) Estimating the impact of different cigarette package warning label policies: the auction method. Addict Behav 32 (12):2916- 2925. doi:S0306-4603(07)00168-2 [pii]
10.1016/j.addbeh.2007.05.018
Ukraine Global Adult Tobacco Survey report (2010). Kiev
White V, Webster B, Wakefield M (2008) Do graphic health warning labels have an impact on adolescents' smoking-related beliefs and behaviours? Addiction 103 (9):1562-1571. doi:ADD2294 [pii]
10.1111/j.1360-0443.2008.02294.x [doi]
WHO (1998) Guidelines for controlling and monitoring the tobacco epidemic. WHO, Geneva, Switzerland
WHO (2009) The current status of the tobacco epidemic in Poland. Copenhagen Willemsen MC (2005) The new EU cigarette health warnings benefit smokers who
want to quit the habit: results from the Dutch Continuous Survey of Smoking Habits. Eur J Public Health 15 (4):389-392. doi:cki061 [pii]
10.1093/eurpub/cki061 [doi]
Tables
Table 1 - Percentage distribution of the study sample by gender, age, education and smoking status – Ukraine omnibus survey on tobacco packs health warnings, 2009
Characteristics Values N %*
Gender Male 861 40,8
Female 1251 59,2
Age 18-24 years 193 9,1
25-34 years 326 15,4
35-44 years 371 17,6
45-54 years 438 20,7
55+ years 784 37,1
Education Incomplete secondary (and less) 296 14,0
Complete secondary 752 35,6
Vocational secondary 585 27,7
Complete higher 479 22,7
Smoking status Daily smokers 533 25,2
Occasional smokers 58 2,7
Reducers 24 1,1
Former daily smokers 195 9,2
Former occasional smokers 36 1,7
Experimenters 119 5,6
Never smokers 1147 54,3
*Weighted data
Table 2 - Number of health warnings recalled by respondent. Results of multivariate analysis performed with the use of General linear model – Ukraine omnibus survey on tobacco packs health warnings, 2009
Characteristics Values of the first variable Values of the second variable B
95%
Confidence
Interval Sig. t Lower
Bound
Upper Bound
Intercept 2,433 2,106 2,760 0,000 14,590
Smoking
status Current smokers 1,061 0,639 1,484 0,000 4,932
Former smokers -0,222 -0,657 0,213 0,318 -1,000
Non-smokers 0,000
ref
group . . .
Age 18-24 years 1,322 0,938 1,705 0,000 6,758
25-34 years 1,216 0,879 1,552 0,000 7,090
35-44 years 0,928 0,607 1,249 0,000 5,668
45-54 years 0,738 0,424 1,051 0,000 4,611
55+ years 0,000
ref
group . . .
Smoking
status x Age Current smokers 18-24 years -0,619 -1,250 0,011 0,054 -1,927
25-34 years -0,323 -0,890 0,243 0,264 -1,118
35-44 years -0,491 -1,033 0,051 0,076 -1,775
45-54 years -0,926 -1,471 -0,382 0,001 -3,336
55+ years 0,000 ref
group . . .
Former smokers 18-24 years 0,504 -0,617 1,624 0,378 0,881
25-34 years 0,453 -0,306 1,212 0,242 1,171
35-44 years -0,576 -1,343 0,190 0,140 -1,475
45-54 years -0,079 -0,739 0,582 0,816 -0,233
55+ years 0,000
ref
group . . .
Education Incomplete secondary (and less) -0,790 -1,174 -0,406 0,000 -4,037
Complete secondary -0,630 -0,930 -0,330 0,000 -4,118
Vocational secondary -0,492 -0,799 -0,185 0,002 -3,139
Complete higher 0,000
ref
group . . .
Gender x
Education Males Incomplete secondary (and less) 0,661 0,182 1,141 0,007 2,705 Complete secondary 0,314 0,020 0,608 0,036 2,094 Vocational secondary 0,218 -0,106 0,543 0,187 1,320 Complete higher -0,322 -0,662 0,018 0,063 -1,858
Table 3 - Frequency of recall of particular textual health warnings and associated factors, stratified by smoking status, results of multivariate binary logistic regression analysis – Ukraine omnibus survey on tobacco packs health warnings, 2009
Health warning 'Smoking
causes cardio- vascular diseases and lung cancer'
'Smoking when pregnant harms your baby'
'Tobacco smoke harms the health of those around you'
'Smoking is addictive, do not start to smoke!'
'Smoking causes impotence'
'Protect children: do not make them
breathe your smoke'
'Smokers die early'
Smokin
g status variable option N %*
OR (95%
CI) %*
OR (95
%
CI) %*
OR (95
%
CI) %*
OR (95%
CI) %*
OR (95
%
CI) %*
OR (95
%
CI) %*
OR (95
% CI)
smoking status
current
smokers 539 87,9 4,59 (3,52 -
6,00) 37,2 1,91 (1,55 -
2,36) 38,3 1,55 (1,26 -
1,91) 24,1 1,73 (1,36 -
2,21) 39,9
3,10 (2,49 -
3,87) 24,0
1,50 (1,18 -
1,90) 43,6
2,02 (1,65 - 2,48)
former
smokers 210 66,2 1,25 (0,93 -
1,69) 19,9 0,81 (0,57 -
1,14) 27,7 0,95 (0,70 -
1,31) 11,3 0,69 (0,45 -
1,07) 19,5
1,13 (0,79 -
1,61) 14,3
0,79 (0,53 -
1,18) 23,8
0,83 (0,60 - 1,15) non-
smokers 1248 61,2 1,00 23,7 1,00 28,5 1,00 15,5 1,00 17,6 1,00 17,4 1,00 27,6 1,00
Current 539
Gender Male 401 87,6 1,00 33,7 1,00 37,9 1,00 24,6 1,00 40,3 1,00 21,8 1,00 43,9 1,00 Female 138 88,4 0,93 50,8 1,55 39,5 0,83 21,9 0,71 38,0 0,79 32,6 1,36 41,9 0,87
(0,49 - 1,77)
(1,02 - 2,36)
(0,55 - 1,28)
(0,43 - 1,15)
(0,52 - 1,21)
(0,87 - 2,14)
(0,57 - 1,33) 5 age
groups 18-24 years 80 92,4 1,00 48,6 1,00 39,0 1,00 24,8 1,00 39,0 1,00 26,7 1,00 51,4 1,00
25-34 years 129 90,6 0,80 (0,32 -
1,97) 46,8 0,90 (0,54 -
1,51) 44,3 1,24 (0,73 -
2,08) 31,4 1,54 (0,87 -
2,74) 53,8
1,90 (1,13 -
3,17) 32,7
1,25 (0,71 -
2,19) 46,5
0,84 (0,50 - 1,41)
35-44 years 117 91,0 0,85 (0,34 -
2,17) 35,2 0,54 (0,32 -
0,92) 36,8 0,90 (0,53 -
1,55) 20,8 0,89 (0,48 -
1,66) 41,0
1,18 (0,70 -
2,00) 24,1
0,85 (0,47 -
1,54) 45,8
0,77 (0,45 - 1,30)
45-54 years 107 85,2 0,45 (0,18 -
1,12) 26,1 0,37 (0,21 -
0,67) 27,8 0,61 (0,34 -
1,10) 16,7 0,70 (0,35 -
1,38) 33,3
0,84 (0,48 -
1,49) 11,3
0,35 (0,17 -
0,74) 33,3
0,43 (0,24 - 0,76)
55+ years 106 77,6 0,27 (0,11 -
0,65) 27,1 0,40 (0,22 -
0,73) 41,1 1,11 (0,62 -
1,96) 25,2 1,12 (0,59 -
2,13) 25,2
0,53 (0,29 -
0,96) 22,4
0,81 (0,42 -
1,54) 39,3
0,61 (0,34 - 1,07) Educatio
n (4 groups)
Incomplete secondary
(and less) 46 84,9 1,00 26,4 1,00 34,0 1,00 15,1 1,00 49,1 1,00 22,6 1,00 39,6 1,00
Complete
secondary 213 88,9 1,16 (0,47 -
2,85) 32,0 1,53 (0,76 -
3,10) 33,2 1,21 (0,63 -
2,32) 23,7 2,05 (0,89 -
4,68) 40,5
0,70 (0,38 -
1,32) 19,4
0,96 (0,46 -
2,00) 44,0
1,25 (0,67 - 2,35) Vocational
secondary 152 89,2 1,19
(0,46 42,0 2,38
(1,16 38,6 1,45
(0,74 21,6 1,72
(0,73 35,2
0,53
(0,28 24,3
1,13
(0,53 40,9
1,25 (0,65
- 3,08)
- 4,88)
- 2,82)
- 4,04)
- 1,03)
- 2,41)
- 2,41)
Complete
higher 128 85,1 0,73 (0,28 -
1,91) 44,2 2,05 (0,99 -
4,27) 48,0 2,14 (1,08 -
4,24) 30,6 2,54 (1,08 -
5,97) 40,8
0,65 (0,33 -
1,27) 32,0
1,52 (0,71 -
3,25) 46,9
1,59 (0,81 - 3,12)
Former 210
Gender Male 154 62,0 1,00 33,7 1,00 24,0 1,00 11,2 1,00 15,7 1,00 10,7 1,00 22,3 1,00
Female 56 80,8 1,49 (0,63 -
3,52) 50,8 1,00 (0,41 -
2,43) 40,4 1,21 (0,56 -
2,64) 11,3 0,53 (0,16 -
1,71) 30,8
0,96 (0,40 -
2,33) 26,4
1,42 (0,51 -
3,94) 28,8
1,03 (0,45 - 2,35) 5 age
groups 18-24 years 11 84,6 1,00 46,2 1,00 42,9 1,00 21,4 1,00 57,1 1,00 28,6 1,00 28,6 1,00
25-34 years 37 81,0 0,87 (0,16 -
4,62) 40,5 0,80 (0,21 -
3,07) 50,0 1,21 (0,34 -
4,32) 19,0 1,61 (0,30 -
8,45) 40,5
0,51 (0,14 -
1,89) 38,1
3,30 (0,70 - 15,5
2) 31,0
1,56 (0,38 - 6,38)
35-44 years 26 76,7 0,77 (0,14 -
4,26) 9,7
0,12 (0,02 -
0,66) 22,6 0,40 (0,10 -
1,62) 3,3
0,21 (0,02 -
2,11) 16,7
0,15 (0,03 -
0,66) 9,7
0,32 (0,05 -
2,02) 22,6
0,85 (0,19 - 3,83)
45-54 years 39 69,8 0,72 (0,14 -
3,69) 16,3 0,27 (0,06 -
1,15) 27,9 0,54 (0,15 -
2,03) 16,7 0,93 (0,17 -
5,06) 18,6
0,15 (0,04 -
0,63) 9,3
0,33 (0,06 -
1,93) 23,3
0,87 (0,20 - 3,72)
55+ years 97 53,9 0,28 (0,06
- 12,6
0,15 (0,04 - 17,6
0,28 (0,08
- 6,9
0,39 (0,07
- 6,8
0,05 (0,01 - 6,9
0,24 (0,05 - 20,6
0,68 (0,17 -
1,34) 0,61) 1,02) 2,11) 0,22) 1,24) 2,73) Educatio
n (4 groups)
Incomplete secondary
(and less) 43 60,5 1,00 16,3 1,00 23,3 1,00 2,3 1,00 11,6 1,00 9,3 1,00 20,9 1,00
Complete
secondary 62 64,8 0,91 (0,37 -
2,28) 11,1 0,43 (0,12 -
1,58) 26,4 0,99 (0,35 -
2,81) 12,5 3,64 (0,47 -
28,23) 18,1
0,89 (0,22 -
3,58) 12,7
0,87 (0,19 -
4,00) 25,0
0,93 (0,33 - 2,65)
Vocational
secondary 54 68,9 0,92 (0,35 -
2,40) 29,5 1,33 (0,42 -
4,27) 27,9 0,80 (0,27 -
2,32) 16,4 4,10 (0,54 -
31,33) 21,3
0,80 (0,20 -
3,13) 21,3
1,43 (0,33 -
6,11) 24,6
0,74 (0,25 - 2,18)
Complete
higher 51 70,9 1,20 (0,45 -
3,21) 23,6 0,88 (0,26 -
2,96) 32,1 0,99 (0,34 -
2,88) 9,1
2,17 (0,26 -
17,85) 25,0
1,11 (0,29 -
4,30) 12,7
0,39 (0,08 -
1,98) 25,5
0,68 (0,23 - 2,02) Non-
smokers 1241
Gender Male 184 61,8 1,00 22,7 1,00 35,5 1,00 21,4 1,00 25,5 1,00 20,0 1,00 35,9 1,00
Female 1057 61,0 1,16 (0,84 -
1,60) 24,0 1,31 (0,91 -
1,91) 26,9 0,76 (0,54 -
1,06) 14,2 0,68 (0,46 -
1,01) 15,7
0,62 (0,42 -
0,90) 16,8
0,96 (0,65 -
1,42) 25,6
0,69 (0,49 - 0,97) 5 age
groups 18-24 years 102 79,2 1,00 38,8 1,00 42,5 1,00 21,5 1,00 32,2 1,00 20,0 1,00 38,0 1,00
25-34 years 160 74,9 0,79 (0,45 -
1,39) 33,9 0,72 (0,43 -
1,19) 42,1 0,89 (0,54 -
1,45) 22,2 0,98 (0,54 -
1,77) 26,3
0,64 (0,37 -
1,11) 29,8
1,61 (0,91 -
2,84) 39,0
0,99 (0,59 - 1,64) 35-44 years 179 68,3 0,57 30,0 0,64 28,3 0,50 17,2 0,81 23,9 0,63 21,7 1,16 30,6 0,77
(0,33 - 0,99)
(0,39 - 1,07)
(0,30 - 0,84)
(0,44 - 1,49)
(0,36 - 1,08)
(0,64 - 2,09)
(0,46 - 1,29)
45-54 years 219 68,4 0,57 (0,33 -
0,99) 24,1 0,47 (0,28 -
0,80) 29,4 0,55 (0,33 -
0,91) 16,0 0,76 (0,41 -
1,40) 18,2
0,46 (0,26 -
0,80) 19,3
0,99 (0,55 -
1,80) 31,6
0,77 (0,46 - 1,29)
55+ years 581 46,2 0,24 (0,14 -
0,39) 13,8 0,21 (0,13 -
0,35) 19,9 0,33 (0,21 -
0,53) 10,7 0,38 (0,21 -
0,69) 8,4
0,17 (0,10 -
0,30) 9,8
0,42 (0,23 -
0,75) 18,4
0,35 (0,21 - 0,56) Educatio
n (4 groups)
Incomplete secondary
(and less) 202 44,0 1,00 19,9 1,00 22,9 1,00 18,2 1,00 14,3 1,00 15,3 1,00 22,9 1,00
Complete
secondary 424 61,9 1,53 (1,03 -
2,28) 19,6 0,69 (0,42 -
1,15) 23,8 0,86 (0,54 -
1,38) 13,9 0,59 (0,34 -
1,01) 12,4
0,52 (0,29 -
0,94) 15,5
0,77 (0,44 -
1,33) 28,6
1,16 (0,72 - 1,86)
Vocational
secondary 338 63,6 1,68 (1,10- 2,56) 23,8
0,89 (0,53 -
1,51) 30,5 1,30 (0,80 -
2,10) 11,3 0,47 (0,26 -
0,85) 17,9
0,86 (0,47 -
1,55) 16,2
0,76 (0,43 -
1,36) 25,1
0,91 (0,55 - 1,52)
Complete
higher 277 68,8 1,85 (1,17 -
2,90) 31,3 1,08 (0,63 -
1,85) 36,8 1,45 (0,88 -
2,40) 20,7 0,89 (0,50 -
1,58) 26,8
1,17 (0,64 -
2,13) 23,3
1,13 (0,63 -
2,03) 32,7
1,22 (0,73 - 2,05) Frequency of recall of particular textual health warnings placed on cigarette packs in Ukraine and factors associated with the probability of recall.
% who recall
* Odds ratios (95% confidence intervals) adjusted for all other factors listed in the left column are shown.
Significant associations are marked with bold typescript.
Table 4 - Bivariate logistic regression of perception of health hazard related to tobacco use by recall of health warnings in male current smokers – Ukraine omnibus survey on tobacco packs health warnings, 2009
Number %* OR** 95,0% C.I.for OR
Lower Upper 'Smoking when pregnant harms your baby' not recalled 395 52,2% 1,000
recalled 233 64,8% 1,689 1,210 2,358 'Smoking is addictive, do not start to smoke!' not recalled 477 54,7%
recalled 151 63,6% 1,445 0,991 2,107
'Protect children: do not make them breathe your
smoke' not recalled 477 54,5%
recalled 151 64,2% 1,499 1,027 2,189
'Smokers die early' not recalled 356 51,1%
recalled 273 64,5% 1,735 1,255 2,397
