Cigarette brand loyalty among smokers in six European countries: Findings from the EUREST-PLUS ITC Europe Surveys

INTRODUCTION This study aims to describe the degree of smokers’ loyalty to a specific brand of tobacco products and the variables related to choosing a specific brand among smokers in six European countries. METHODS A cross-sectional analysis was conducted for a representative sample of adult smokers from Germany, Greece, Hungary, Poland, Romania, and Spain (approximately 1000 smokers per country). The prevalence of smokers’ having a usual brand of cigarettes smoked (factory-made or roll-yourown cigarettes), the brand of choice, the factors for choosing a specific brand and the degree of loyalty to that brand (not at all, a little, somewhat and a lot) were assessed by country, sociodemographics and smoking-related variables. RESULTS In total, 86.6% of the smokers reported having a usual brand. In three out of the six countries, one brand holds the loyalty of between 17.8% and 24.5% of the smokers that reported having a usual brand for factory-made cigarettes. Most participants reported being loyal ‘a lot’ to their brand of choice (44.4%). The reasons most reported for choosing a cigarette brand were the taste (83.2%) and the price (51.7%). CONCLUSIONS Brand loyalty is high among factory-made and roll-your-own cigarette smokers in six European countries. Future research on longitudinal trends of brand loyalty to evaluate the effect of tobacco control policies in these European countries is warranted.


INTRODUCTION
Tobacco was the cause of the death of around seven million people in 2017 1 , with an alarming rate of up to two deaths per three smokers 2 . Although it is such a lethal product, tobacco companies still achieve high success in marketing and selling 2 tobacco. In fact, the six largest cigarette companies made a profit of approximately $10000 per death from tobacco smoking, in 2015 3 . Understanding the factors influencing smokers' brand choice and their loyalty to a brand are important as countries develop and implement tobacco control strategies 4 , but there is still scarce academic research in this regard. The European Union (EU) has become an increasingly restrictive market for tobacco products in recent years with the signature of the World Health Organization (WHO) Framework Convention on Tobacco Control (FCTC) 5 , the first public health treaty aiming to tackle some of the causes of the tobacco epidemic, and the approval of a new EU Tobacco Products Directive (TPD) (2014/40/EU), which aims to regulate several aspects surrounding tobacco products 6 . Both of the abovementioned regulations could affect whether and why smokers are loyal to a cigarette brand, although none of these regulations includes measures designed to directly influence brand loyalty.
Diminishing brand loyalty might be an effective pathway to tobacco control, considering brand loyalty has been found to be associated with a decline in the sense of identity smokers may have of being a smoker of a certain brand, and therefore sharing special characteristics with those that share the same brand of choice 7,8 . Such brand identity has been used by the tobacco industry as a means to maintain sales growth. This became clear with the disclosure of tobacco industry internal documents, which have shown that tobacco companies develop explicit marketing strategies to create meaningful identities through tobacco brands 9 . Similarly, there is theoretical support for the link between brand identity and the deflection of bad connotations and stigmatization associated with smoking 10,11 . The implementation of cigarette standardised packaging design was found to be associated with decreased brand identity, decreased positive brand stereotypes, decreased smoking behaviour and higher odds of quitting among smokers 8 .
This paper documents several aspects of tobacco brand loyalty, such as the degree of brand loyalty and variables associated with brand choice of tobacco products among smokers in Germany (DE), Greece (GR), Hungary (HU), Poland (PL), Romania (RO), and Spain (ES), countries where there are still no data about such aspects of brand loyalty.

Study design
The EUREST-PLUS project aims to evaluate the implementation and impact of the TPD and WHO FCTC in six European Member States (MS). The cross-sectional data analysed here derive from Wave 1 of the EUREST-PLUS International Tobacco Control Policy Evaluation (ITC) survey, a longitudinal cohort study of smokers in DE, GR, HU, PL, RO, and ES 12,13 . The fieldwork for Wave 1 was conducted between June 2016 and September 2016 14 . The survey sample comprised 6011 (about 1000 per country) nationally representative adult (over 18 years old) cigarette smokers that had smoked more than 100 cigarettes in their lives. Further details including study design and recruitment can be found elsewhere 12,13 .

Ethics
This study's protocol was approved by the Office of Research Ethics at the University of Waterloo (Canada), and by the ethics committees of all the participant countries and partnering institutions. The EUREST-PLUS Project is registered in Clinicaltrials. gov (registration number NCT02773836).

Brand loyalty measures
Participants were asked: 'Do you have a usual brand and variety of cigarettes?'. Response options were 'yes' and 'no'. If they answered 'yes', they were asked the following questions to explore the degree of brand loyalty: 'To what extent are you committed to your regular brand of cigarettes?'. Answer options were 'not at all', 'a little', 'somewhat', 'a lot', 'don't know', and 'refused'. Those that answered 'don't know' and 'refused' were excluded from the analysis of the degree of loyalty and factors influencing the decision to choose a specific brand.
In addition, respondents were asked: 'What is your usual brand of factory-made cigarettes?' and 'What is your usual brand of roll-your-own cigarettes?' Two lists of brands of roll-your-own (RYO) and factorymade (FM) cigarettes had been previously developed for each country and the answers provided by the respondents were compared to those lists and 3 categorized accordingly; if the brand mentioned was not in the list, the response was categorized as 'other brand'. Participants could also respond 'don't know' or 'refused' and those who chose one of these two options were excluded from the analysis.

Reasons for brand loyalty
Respondents were also asked: 'In choosing your usual brand, was part of your decision to smoke this brand based on any of the following: It may not be bad for your health? The price? How they taste?
The look and feel of the pack? The tar and nicotine levels of the brand?'. The response options for each question were 'yes', 'no', 'don't know', and 'refused'. Those who answered 'don't know' or 'refused' were excluded from the analysis.

Analysis
Descriptive analyses were conducted to compute the prevalence of brand loyalty, the degree of loyalty (not at all, a little, somewhat, and a lot loyal), and the reasons influencing the decision for choosing their usual brand among smokers by country, sociodemographic and smoking-related variables. Pearson's chi-squared test was used to assess differences among groups. The overall most popular brands of RYO and FM cigarettes were identified by country. Additionally, multivariate logistic regression was used to compute adjusted odds ratios (AOR) and 95% confidence intervals (CI) of the association of having a usual brand by sociodemographic variables. All the analysis incorporated the weights from the complex sampling design. Stata version 13 was used for the analyses.

Usual brand
Most respondents reported having a usual brand (86.6%).

Brands most used across countries
The most popular brand in the overall sample was Malboro (14.65%). In three out of the six countries, Marlboro was the usual brand for between 17.8% and 24.5% of smokers that reported having a usual brand for FM cigarettes, and it was the most popular in four of the six countries. The second most popular brand, Kent, reaches this position because of RO smokers (23.3%), but was not highly reported as the brand of choice in the other five countries. RO participants were the ones reporting the most disperse results, with 76.7% of the smokers being loyal to brands other than the most popular ones in the overall sample. Brand loyalty for RYO was more dispersed, with Marlboro being the only usual brand for more than 10%, in two out of the six countries.  Table 3 presents the reasons that may influence the decision for loyalty to a specific brand by sociodemographic variables. Overall, the taste from the cigarette was the reason most cited by smokers (83.2%) followed by the price (51.7%), the tar and nicotine levels of the brand (41.5%), the look and the feel of the pack (23.2%), and the perception of being less harmful (21.4%).

Reasons for choosing usual brand
Female smokers were significantly more influenced than male smokers by the price (53.2% vs 50.5%), the tar and nicotine levels (44.5% vs 39.2%), the look and feel of the pack (25.5% vs 21.4%), and the cigarette harm perception (23.8% vs 19.6%). The younger the smokers the bigger the percentage of them reporting

DISCUSSION
Overall, most smokers (86.6%) reported having a usual brand, particularly daily smokers (87.4%) and those highly educated (90.2%). One brand was the most popular in four out of the six countries, holding the loyalty of less than a quarter of the smokers in these countries. Almost half of the smokers reported being loyal 'a lot' to a brand and only 5.1% reported having the lowest level of loyalty. Taste was reported by eight in ten as a factor that influences their loyalty, and five in ten mentioned price as an influential factor. Smokers with the lowest educational level were less brand loyal and reported price as the main influential factor. Other studies have found similar results, with smokers with lower socioeconomical levels switching brands more frequently because of price differences, i.e. being less brand loyal 15,16 .
One of these studies also found similar results in relation to level of nicotine addiction, with those more addicted to nicotine being more brand loyal than those less addicted 16 . One possible explanation for these results is that those less addicted are mostly occasional smokers and tend to share cigarette packs with other smokers making their choice more defined by the others. Gender differences were observed in reporting brand loyalty and factors influencing it, with females being more frequently loyal to a brand and mentioning price, the levels of nicotine and tar, the look and feel of the pack, and the perceived potential harms as factors influencing their choice of a cigarette brand more than males. These results might be related to the fact that in the EU, in 2016, the gender pay gap (the average difference between the renumeration for men and women working) was 16.2% and all the countries in the study the gender pay gap was at least 15.3%, except in RO (5%) 17 . Evidence also presents women as more attentive to health-related issues and more likely to take action on these matters [18][19][20] . Additional evidence related to brand loyalty and gender differences, that might at least partially explain our results, is that the tobacco industry has systematically targeted women in their advertisements, associating brands with positive images of independence, self-care and success 16,21,22 .
Marlboro, the most popular choice of usual brand overall, was also the most popular choice for RYO and FM smokers in four out of the six countries. Marlboro was the most sold cigarette brand in 2016 in DE, GR, HU, and ES; with DE being the country with the most affordable Marlboro cigarettes, followed by ES, GR, and HU 23-26 . As the price was the second most cited reason for choosing a brand (51.7% of the sample), the increase in taxation of cigarettes could play a part in the decrease of brand loyalty in these countries.
A difference of approximately 20% prevalence was observed among the countries with the highest and lowest brand loyalty prevalence. This finding might point to cross-country differences in the implementation of tobacco control policies, such as bans on tobacco advertising. The EUREST-PLUS ITC 6E Survey shows that tobacco advertising, promotion and sponsorship tended to be noticed more often in countries with less restrictive regulation (e.g. DE and GR) 27 . Another finding from the same EUREST-PLUS ITC 6E Survey is that tobacco advertising exposure was widely prevalent outside and inside the points-of-sale in all six countries, including HU and RO, countries where bans on this advertising have been implemented 27 .
Taste of the cigarette was the factor most cited as a reason for choosing a brand by 8 in 10 smokers, followed by price (approximately 5 in 10 smokers), tar and nicotine levels for brand (4 in 10 smokers), look and the feel of the pack (2 in 10 smokers), and perception of harms (2 in 10 smokers). Data from the Eurobarometer also indicated that just under half of the smokers in the EU smoked cigarettes with special characteristics in 2017 28 . The most popular choices were the additive-free or organic cigarettes (17%) and the light cigarettes (16%). Our data point to the possibility that most smokers in Europe smoke the cigarettes they perceive as less harmful, even though there is no evidence for differences in the safety of combustible cigarettes 29,30 . Additionally, menthol flavour (8%) and other flavours (7%) were also identified in the Eurobarometer survey as consistently used by smokers. Therefore, these results reinforce the importance of the effective ban of flavourings in cigarettes established by the TPD, which should be fully implemented by 2020 6 .
Another measure that has been found to be related to brand loyalty is plain packaging 8 . It aims to standardise and eliminate the design and packaging characteristics that could mislead consumers, suggesting benefits in terms of less harm. Several studies provide evidence that it may reduce false beliefs that certain brands are less harmful and reduce pack and product appeal 9,31,32 . At the population level, plain packaging has already been proven to be effective in tackling tobacco consumption by lowering brand loyalty in Australia 8 . Additionally, Australian smokers were more likely to find their tobacco product packs less attractive, find their cigarettes less satisfying, and consider cigarette brands not differing in prestige, after one year of plain packaging implementation 8 .
In the EU, most of the measures transposed and implemented were not designed to undermine the relationship of smokers with their specific brand, which could act as a shield against tobacco cessation 9 . Meanwhile, measures such as plain packaging, that could lead to a direct change in brand loyalty, are not compulsory and were left in charge of each country by the TPD 33 . This may result in delays and interference of the tobacco industry in the process of adopting plain packaging 33 .
Within the participating countries of the EUREST-PLUS Project, HU is going to implement plain packaging in 2018, which will enable a crosscountry comparison of the effects of such a measure in a longitudinal analysis. Longitudinal analysis of the EUREST-PLUS ITC Europe Survey will provide a clearer assessment of the potential association between the plain packaging policies implemented in some of the participating countries and changes in brand loyalty. Longitudinal analysis will also allow us to explore associations between brand loyalty and smoking cessation outcomes.
This study has some limitations, which should be noted. The pre-specified answer options for the questions might potentially exclude some important elements that influence brand choice and loyalty among smokers. The data used in this study are crosssectional and this design precludes any inference about the direction of causality. The question evaluating the reasons that influence a smokers' brand choice could be interpreted in two ways: 1) smokers choose their cigarettes because of their higher content of nicotine and tar, and 2) smokers choose their cigarettes because of their lower content of nicotine and tar as a healthier option. Although it seems intuitive they would choose the cigarettes with less concentrations of these compounds, the question could have been clearer. Some strengths should also be noted. This study used nationally representative samples of smokers in each of the six countries, making the generalisation of these results more feasible. Also, a standardised questionnaire and methodology were used for all the countries included in the current study, thus assuring comparability across countries.

CONCLUSIONS
This study provides a snapshot of cigarette brand loyalty across six European countries for the first time. The measures used here suggest that brand loyalty is high among smokers in these EU countries.

PROVENANCE AND PEER REVIEW
Commissioned; externally peer reviewed.