Article Text

Download PDFPDF

Original research
Influence of cigarette packet branding and colours on young male smokers’ recognition, appeal and harm perceptions of tobacco brands in Cambodia: a mixed-methods study
  1. Thomas Stubbs1,
  2. Victoria White1,
  3. Hua-Hie Yong1,
  4. Chhea Chhordaphea2,
  5. John W Toumbourou3
  1. 1School of Psychology, Faculty of Health, Deakin University, Burwood, Victoria, Australia
  2. 2National Center for Health Promotion, Ministry of Health, Phnom Penh, Cambodia
  3. 3Centre for Drug use, Addictive and Anti-social behaviour Research, School of Psychology, Faculty of Health, Deakin University, Burwood, Victoria, Australia
  1. Correspondence to Thomas Stubbs; tstubbs{at}; Professor John W Toumbourou; john.toumbourou{at}


Objectives To explore how cigarette packet branding and colours influence young male smokers’ perceptions of tobacco brands in Cambodia.

Design Mixed-methods study.

Setting Worksites, living accommodations, a university and public locations in Phnom Penh, Cambodia.

Participants 147 male Cambodian smokers (18–24 years).

Interventions Participants were shown mock-up pictures of different cigarette packet branding and colour variations and asked to respond to close-ended and short-response questions.

Outcome measures Brand recognition, appeal and harm perceptions of cigarette packet branding and colours.

Results When shown three packets with brand names removed, 98.6% of participants recognised packet one as Mevius brand, 21.1% recognised packet two as Marlboro and 38.8% recognised packet three as 555. For the three fully-branded and three matching plain packets, most participants selected a fully-branded packet as the most appealing taste (83.0%) and most appealing to youth (81.7%). Participants described their chosen brand as appealing due to beliefs about its superior taste/quality, reduced harm and symbolic attitudes surrounding tobacco brands and smokers of different brands in a social status hierarchy. When shown six different colours of unbranded packets, participants selected the blue packet (51.0%) as the most appealing for taste, the white packet as the least harmful (25.2%), and the red (15.0%) and black (12.9%) packets as the most harmful to health. They described their associations of packet colours with abstract imagery concerning smoking-related harms and their future well-being.

Conclusions Findings suggest that packet branding and colours influence young male smokers’ recognition, appeal and harm perceptions of tobacco brands in Cambodia and remain an influential marketing tool for tobacco companies where advertising is banned. Consequently, Cambodia and other low and middle-income countries in Southeast Asia should implement plain packaging.

  • public health
  • qualitative research
  • health policy

Data availability statement

No data are available.

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See:

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Strengths and limitations of this study

  • This is the first study to explore the influence of cigarette packet branding and colours in Cambodia.

  • The survey was pretested with a sample of young male smokers in Cambodia to ensure the questions were linguistically and culturally appropriate.

  • Data were collected by Cambodian research assistants and in the local language of participants.

  • Participants were recruited using non-probability, convenience sampling, hence the results may not generalise to the broader population.


In response to restrictions on tobacco advertising, promotion and sponsorship (TAPS), tobacco companies turned to cigarette packet branding to advertise their brands.1 2 This marketing tool—which includes brand elements on product packaging such as logos, colours, images and descriptions—provides tobacco companies with various opportunities to reach current and potential smokers.3 For example, individuals are exposed to brand elements when cigarette packets are displayed in retail stores or when smokers publicly display their packets while retrieving a cigarette or during smoking.3 4 Consequently, smokers become a ‘silent salesperson for tobacco companies, unknowingly exposing others to brand elements.5

Packet brand elements have been shown to influence attitudes and consumption behaviours, with branded packets perceived as more appealing and having better quality cigarettes than plain packets.6–8 Packet branding may also influence the appeal of tobacco brands through brand imagery, with young people describing tobacco brands that used appealing images on their packets as ‘mature’ and ‘sophisticated’.9 Packet brand elements may also influence perceived characteristics of typical smokers of the brand, with brand names, descriptions, colours and images associated with ‘masculine’ or ‘feminine’ brands.6 10 Qualitative studies also suggest that young people use tobacco brands as a way to define and communicate their self-image to others and elevate their social status.11 12

The tobacco industry has a history of using marketing tactics to create the impression that certain tobacco products are less harmful than others.13 Literature suggests that packet branding may influence young people’s harm perceptions about tobacco products, with brand descriptors (such as ‘light’, ‘smooth’ or ‘additive free’) potentially creating an impression of reduced harm.14–17 Analysis of tobacco industry documents has shown that companies used packet colours to influence consumers’ perceptions about the taste, strength and health impacts of cigarettes.18 Research suggests that smokers associate lighter coloured packets with cigarettes that are weaker in strength, contain fewer harmful substances and are less harmful, while darker-coloured packets are associated with increased harm and reduced appeal.19 20

Low and middle-income countries (LMICs) in Southeast Asia have made substantial efforts to reduce tobacco use, with most ratifying the World Health Organization’s (WHO) Framework Convention on Tobacco Control (FCTC) and introducing comprehensive bans on TAPS.21 However, most of these countries still permit cigarette packet branding, with evidence showing that tobacco companies use packet branding as a marketing tool in the region.22 23 In the Philippines, one qualitative study showed that young people describe flavoured cigarette packets as attractive and associated the colour of those packets with product harm—often perceiving lighter colours with reduced harm and red or darker colours with increased harm.24 With data showing that smoking uptake mainly occurs among men during adolescence and young adulthood,21 there is a lack of evidence from LMICs in Southeast Asia to explain how packet branding impacts young people’s smoking attitudes and consumption behaviours, or how plain packaging might reduce the effectiveness of packaging as a marketing tool.25

Cambodia is a lower middle-income country in Southeast Asia that continues to face a significant health and economic burden from tobacco use. An estimated 15 000 Cambodians die each year from tobacco-related illnesses, with annual costs reaching 3% of the country’s gross domestic product.26 Like other countries in the region, smoking is a male-dominated practice in Cambodia—with data showing that men account for more than 90% of the country’s 1.68 million smokers.27 Similarly, smoking uptake predominately occurs during adolescence to young adulthood in Cambodia.27 The Cambodian government has introduced TAPS restrictions but still permits marketing through cigarette packet branding.28 To our knowledge, no study has explored how the tobacco industry has used cigarette packet branding as a marketing tool in Cambodia, or how packet branding might influence young people’s smoking and brand attitudes.

This study explored how cigarette packet branding (such as logos, descriptions and colours) influence the recognition, appeal and harm perceptions of tobacco products among young male smokers in Cambodia. Three questions guided the research:

  1. To what extent can young male smokers recognise specific tobacco brands from cigarette packets with brand names removed?

  2. Is there evidence that cigarette packet branding influences the appeal of, and attitudes towards, different tobacco brands or products?

  3. Is there evidence that cigarette packet colours influence the appeal and harm perceptions of tobacco brands or products?



This research used a mixed-methods design, which collected quantitative and qualitative data to obtain insights into participants.29 This research was part of a broader study that explored young male smokers’ attitudes and experiences surrounding combustible cigarette smoking, tobacco marketing and e-cigarette use in Cambodia.30 These topics were presented to participants in discrete sections to avoid any potential confusion or cross-influence in responses between combustible cigarette smoking and e-cigarette use.

Sample and recruitment

In early 2020, we recruited a convenience sample of approximately equal numbers of young male smokers from worksites, living accommodations, a university and public locations in Phnom Penh, Cambodia—enabling insights into participants of diverse socioeconomic backgrounds to be obtained. Construction, tourism and garment manufacturing were the targeted workplaces reflecting the country’s leading employment sectors.31 We planned to recruit approximately 150 individuals to reach saturation for the qualitative, short-response questions in the survey.32 Eligibility criteria included: Cambodian citizenship, male, 18–24 years and current smoker (defined as having smoked at least one cigarette in the past month). We provided individuals with the study’s plain language statement and consent form in Khmer. Participants provided written consent and received a US$3 phone voucher for their time.

Data collection

We designed a mixed-methods, interview-administered survey, which included concepts and questions based on previous studies that explored young people’s attitudes towards cigarette packet branding and colours.14 15 17 19 20 The survey was translated from English to Khmer by professional translators and pretested with a sample of young male smokers in Cambodia, which led to minor revisions to ensure that the questions were linguistically and culturally appropriate. We trained and supported local research assistants (two were bilingual in English and Khmer) on data collection protocols and to conduct surveys. The surveys were implemented using digital tablets installed with Qualtrics, with participants selecting their language preference (Khmer or English). Close-ended questions obtained data on participants’ demographic characteristics and smoking behaviours. Short-response questions were used to collect qualitative data, which were recorded verbatim in text format. Interviewer probes were used to illicit additional details. The surveys took approximately 30 min to complete for each participant.

Section 1. Packet brand recognition

To explore brand recognition, we showed participants a picture of three cigarette packets sold in Cambodia with brand names removed (figure 1). These three brands were identified in a previous study as having appealing packet branding in Cambodia and neighbouring countries.23 Using open-ended questions, we asked participants if they could name the tobacco brand of each packet.

Figure 1

Packets with brand names removed.

Section 2. Packet brand appeal and attitudes

We showed participants a picture of three fully-branded packets of cigarettes sold in Cambodia and three matching plain packets, which include brand names printed on neutral, plain grey packets with no brand elements (figure 2). All six packets included graphic health warnings (GHWs) used in Cambodia at the time of this study. We asked participants to select the packet they thought would be the most appealing in terms of taste and most appealing to young people in Cambodia. Participants then provided short responses to describe why they thought their chosen packet was the most appealing. Responses were recorded verbatim in text.

Figure 2

Fully branded packets and matching neutral plain packets.

Section 3. Packet colour appeal and harm perceptions

We then showed participants a picture of six unbranded cigarette packets. Each packet was a different colour with no brand name or brand elements (figure 3). Each packet included a GHW used in Cambodia at the time of the study. The packets included a mix of lighter and darker colours as well as colours used in the branding of tobacco brands sold in Cambodia (Marlboro, Mevius, ARA). Participants selected the packet they thought would be the most appealing in terms of taste, least harmful and most harmful. Participants then provided short responses to describe these choices, which were recorded verbatim in text.

Figure 3

Unbranded packets in different colours.

Data analysis

We used descriptive statistics (using SPSS V.27) to assess participants’ demographic characteristics, regular cigarette brand smoked, packet brand recognition, cigarette packet brand appeal and packet colour appeal and harm perceptions. We collapsed participants’ regular cigarette brand smoked into a binary variable of ‘Mevius’ smokers (the most commonly smoked brand) and all other brands (combined to ensure sufficient sample size for comparison). We used χ2 analyses to examine whether responses to packet brand recognition, packet brand appeal, packet colour appeal and harm perceptions differed between participants who smoked ‘Mevius’ and those who smoked the other brands. For analyses with small cell sizes, we used the Fisher’s exact test to examine differences.

The qualitative data were translated from Khmer to English by two professional translators, with minor grammatical errors corrected. We then analysed the qualitative data using inductive, reflexive thematic analysis. This involved data familiarisation, data coding and arranging codes into themes and subthemes.33 Field notes were used to embed contextual information throughout this process.34 We practised reflexivity to consider how our sociocultural backgrounds might have impacted our data analysis and interpretation.35 Potential factors that may have influenced data analysis included our nationality and experience working in Cambodia and other LMICs across the region. The authors discussed the analysis to promote reflexivity, which led to refinement of qualitative themes.


Participant characteristics

One hundred and forty-seven young men participated (table 1). Most participants were aged 21 years or younger (n=83, 56.5%), over one-third had not completed secondary school (n=63, 42.9%), and around half earned more than US$200 per month (n=80, 54.4%). Most smoking was at a relatively low level, with half smoking an average of five or less cigarette per day (n=82, 55.7%). Around three-quarters of participants indicated ‘Mevius’ as their regular cigarette brand smoked (n=112, 76.2%).

Table 1

Demographic characteristics

Section 1. Packet brand recognition

Nearly all participants correctly recognised the brand name of packet one as Mevius (n=145, 98.6%), approximately one-fifth recognised packet two as Marlboro (n=31, 21.1%), and one-third recognised packet three as 555 (n=57, 38.8%) (figure 1). Nearly half correctly recognised the brand on two or more packets (n=66, 44.9%), while one in seven participants recognised all three brands (n=22, 14.9%). There was no significant difference between participants who smoked ‘Mevius and those who smoked other brands for recognition of the three brands (all p values >0.05).

Section 2. Packet branding appeal and attitudes

Most participants selected a fully-branded packet as the most appealing in terms of taste (n=122, 83.0%), with a similar proportion selecting a fully branded packet as the most appealing to youth (n=120, 81.7%) (figure 2) (table 2). Around one-tenth of the participants thought that there was no difference between all the packets regarding taste (n=16, 10.9%) or appeal to youth (n=18, 12.2%). More participants selected the Mevius plain packet as the most appealing taste (n=7, 4.7%) and to youth (n=4, 2.7%) than the ARA or Marlboro plain packets. Brand smoked was not related to brand selection for taste appeal (p=0.166) or youth appeal (p=0.265).

Table 2

Appeal of fully branded packets and matching plain packets

Qualitative data on packet branding appeals (taste and youth) revealed four themes, with specific quotes for each theme presented in table 3. The influence of participants’ familiarity with some brands was noted in each of the themes.

Table 3

Participants’ quotes demonstrating themes relating to packet branding appeal, perceptions, and attitudes

Theme 1. Brand appeal

When describing why they selected a branded packet, participants commonly mentioned the attributes of the cigarettes. Two main subthemes were found: quality and taste appeal and attractiveness of flavour change capsules. The first subtheme indicates that participants saw the brand selected as providing superior quality and tasting cigarettes, describing them as ‘the best quality, ‘fragrant’, ‘very delicious’ or as having a ‘nice smell’. Some participants noted that the lighter smell and taste of these brands were particularly appealing to young people. For the second subtheme, some participants noted that the branded packets were appealing because they offered cigarettes with flavour change capsules, which could be used to change the taste of the smoke to ‘menthol’, ‘fruit’, ‘grape’ or ‘chocolate.

Theme 2. Harm perceptions

Harm perceptions were influenced by participants’ beliefs about the strength of and ease of smoking the cigarettes. Participants indicated that some branded packets were weaker than others, describing them as ‘light, ‘not too strong or ‘easy to smoke. One participant likened this weaker smoke to the aerosol produced from e-cigarettes, claiming it was ‘like smoke from vaping… sort of light’. Participants also associated weaker cigarettes as being less harmful than other brands, stating that they contained ‘less nicotine’, were ‘less addictive’ and caused less damage to their ‘lungs’ or ‘throat’. Some participants noted that the weaker cigarettes were suitable for youth or those trying smoking for the first time, while others described these cigarettes as suitable for smokers who are trying to reduce their smoking.

Theme 3. Social hierarchies

When describing branded packets, participants associated tobacco brands with different socioeconomic groups. For example, participants often distinguished between the different brands smoked by young people and those smoked by older people or the different brands smoked by those in cities and rural residents. Some participants indicated that smoking the youth brand elevated one’s social status, with one participant stating that young people smoked this brand to ‘show off to other people’. Others stated that smoking the youth brand meant that they could engage in social smoking with other young people who smoked this brand, like sharing cigarettes within their peer groups.

Theme 4. Brand loyalty

Some participants appeared to select branded packets out of a sense of loyalty to that brand, often recalling it as familiar or identifying it as the brand that they regularly smoked. For example, participants indicated their preference for their brand over other alternatives, with one participant claiming that ‘I just smoke Mevius, the brand I know’. Some stated that they consistently smoked the same brand of cigarettes and were hesitant to change, with one participant claiming that ‘I cannot change my brand’. Moreover, there was evidence that some participants developed a sense of connection with their brand during smoking initiation, stating that they had been smoking ‘this brand from the beginning’.

Section 3. Packet colour appeal and harm perceptions

More than half of the participants selected the blue cigarette packet as the most appealing for taste (n=75, 51.0%) and nearly 8% selected the white packet (n=11, 7.5%) (figure 3) (table 4). When harm was considered, 25.2% (n=37) selected the white cigarette packet as the least harmful to health and 10.2% (n=15) selected the blue packet. In contrast, 15% (n=22) selected the red packet and 12.9% (n=19) selected the black packet as the most harmful. Between participants who smoked ‘Mevius’ and those who smoked all other brands, there was no significant difference in the selection of packets in terms of taste (p=0.071), least harmful (p=0.464) or most harmful (p=0.445).

Table 4

Packet colour appeal and harm perceptions

Two themes emerged regarding packet colour appeal and harm perceptions. Specific quotes for each theme are presented in table 5.

Table 5

Participants’ quotes demonstrating themes relating to packet colour appeal and harm perceptions

Theme 1. Packet colour and product appeal

Packet colour was associated with the appeal and flavour of the underlying cigarette, with comments suggesting that cigarettes from blue packs tasted ‘light’ or ‘kind of soft’, while others associated it with ‘menthol’-flavoured cigarettes. In contrast, red and darker coloured packets were associated with stronger flavoured cigarettes. Participants also associated packet colours with tobacco brands, relating the blue packet with the tobacco brand ‘Mevius’ and the red packet with ‘ARA’ or ‘Marlboro’. These associations influenced the appeal of products, with some stating that ‘it looks the same as Mevius, so I think it would taste good’. They also displayed a sense of familiarity and preference with packet colours similar to their current tobacco brand, stating that ‘I kind of already know it or ‘I feel like I already know this taste’.

Theme 2. Packet colour and harm perceptions

Comments regarding packet colour and harm indicated that participants often associated lighter coloured packets (such as blue and white) with cigarettes that had fewer harmful substances. Participants also associated lighter coloured packets with abstract, positive imagery (such as ‘purity’ or ‘a good heart’), which indicated a less harmful product. In contrast, black and red packets were associated with stronger cigarettes, which were seen as most dangerous to smoke. Participants also described black and red packets with negative imagery such as ‘death and danger’, ‘blood’ or an ‘accident in the car’. They perceived this imagery as negative consequences for smoking cigarettes in these packets.


The research showed that, regardless of the brand smoked, most young male smokers recognise the tobacco brand on altered cigarette packets with the brand names removed. With mass media TAPS banned in Cambodia,26 these findings indicate that packet branding still provides tobacco companies with a marketing tool to generate brand awareness among young smokers. This finding is concerning given tobacco companies’ successful use of cigarette packet branding as a marketing tool in high-income countries prior to packet advertising restrictions being introduced.1 2

Quantitative findings showed that most participants selected a specific fully-branded packet as the most appealing in taste and to young people. Participants qualitatively described this branded packet as having a superior taste and quality (often citing flavour varieties and flavour change capsules) or because they perceived it as less harmful than other brands. While studies in high-income countries have demonstrated that individuals perceive branded packets as more appealing6–8 and less harmful than plain packets,14–16 this is the first study, to our knowledge, to demonstrate these same perceptions in young people in Cambodia. This finding aligns with previous studies in the region, which showed that cigarette packets include brand elements and flavours that may appeal to young people.23 24

Analyses of responses to open-ended questions revealed that young Cambodian men also held symbolic attitudes towards tobacco brands, such as believing that a particular brand was suitable for young, higher class smokers while other brands were associated with older, lower class smokers. Findings also showed that participants perceived consumption of certain brands as a way to identify and interact with specific social groups—particularly young, higher class smokers. These findings align with studies in high-income countries, which demonstrated that young people associate tobacco brands with symbolic imagery and user profiles of smokers6 9 10 and consume tobacco brands to communicate and elevate their social status.11 12 Taken together, this evidence shows that cigarette packet branding influences sophisticated brand attitudes among young people—beyond just communicating brand names and product attributes.

Similar to research findings from high-income countries19 20 and the Philippines,24 this study found that young male smokers in Cambodia associated lighter coloured packets (blue and white) with less harmful cigarettes and red and black packets with more harmful cigarettes. The qualitative data also revealed that some participants associated blue-coloured packets with menthol flavoured cigarettes. While menthol cigarettes were commonly associated with green packets in high-income countries,19 our findings align with a recent Filipino study, which showed that menthol cigarettes were also promoted in blue packets, and young people perceived them as less harmful than menthol cigarettes in green packets.24 These findings suggest that tobacco companies may have modified their traditional colour schemes in Southeast Asian LMICs, and that these colours may contribute to further reduced harm perceptions among young people. Moreover, the qualitative data revealed that participants often associated packet colours with abstract imagery concerning smoking-related harms—with lighter colours associated with safer cigarettes and darker colours and red associated with danger and negative outcomes to their well-being. This novel finding provides insights into how young Cambodians perceive and associate packet colours with harm perceptions, often in abstract and emotionally driven ways.

This research has important policy implications. Article 11 of the WHO FCTC requires countries to implement measures to ensure that packaging does not promote tobacco products in any ways that are false, misleading, deceptive or likely to create the false impression of reduced harm compared with other products.36 To this end, the Guidelines on Implementation of Article 11 call for countries to introduce plain packaging legislation that prohibits brand logos, colours, images or promotional information on packaging and requires brand names to be displayed in a standard colour and font.36 Experimental and longitudinal studies in high-income, Western countries demonstrate that plain packaging is effective in reducing the attractiveness of tobacco products and brand imagery.37–39 While all Southeast Asian countries have implemented some restrictions on mainstream forms of TAPS, only Singapore and Thailand have introduced plain packaging (Myanmar planned to introduce this legislation in 2022).21 Consequently, Cambodia and other LMICs in the region should introduce plain packaging to restrict tobacco companies from using this salient and influential marketing tool to influence young people.


The research findings should be considered alongside important limitations. First, all participants were current cigarette smokers. Therefore, they would possibly smoke a regular brand of cigarettes and would have pre-existing attitudes towards that brand, which may have impacted their recognition, appeal and perceptions of the tobacco brands used in this study. This influence was noticed in the qualitative responses; however, the quantitative analysis did not find any association between participants’ regular cigarette brand and their brand recognition and attitudes. Second, a limited number of tobacco brands and colours were displayed on the packets used in this study, so these packets do not reflect all the tobacco brands and colours sold in Cambodia. Further research is required to explore a broader range of packet colours, including potentially new colour schemes concerning menthol or flavoured cigarettes. Third, convenience sampling was used to recruit participants, so the results cannot be generalised to all young male smokers or non-smokers in Cambodia. Fourth, only men were included in the research. While smoking is largely male dominated in Cambodia, these findings should not be used to explain how women may perceive cigarette packet branding directly. Fifth, despite efforts to recruit participants from different sources and diverse socioeconomic backgrounds, data were collected from a non-probability, convenience sample of young male smokers. Therefore, the findings may not generalise to the broader population. Sixth, this research only collected cross-sectional data. Therefore, experimental and longitudinal studies are needed to explore how packet branding, or implementation of plain packaging, may influence young Cambodians’ smoking susceptibility, uptake and consumption.


This study demonstrated that cigarette packet branding and colours influence young male smokers’ recognition, appeal and harm perceptions of tobacco brands sold in Cambodia. Despite increased TAPS restrictions, this research shows that cigarette packet branding and colours still provide tobacco companies in Cambodia with an influential marketing tool to reach young people. This research supports calls for Cambodia and other LMICs in Southeast Asia to implement plain packeting legislation according to Article 11 of the FCTC.

Data availability statement

No data are available.

Ethics statements

Patient consent for publication

Ethics approval

This research was approved by the Deakin University Human Research Ethics Committee [2019-353] and the National Ethics Committee for Health Research of the Ministry of Health in Cambodia [277]. Participants gave informed consent to participate in the study before taking part.


The authors thank the participants for taking part in the research, the research assistants for their help with recruitment and data collection, and the National Center for Health Promotion of the Ministry of Health, Cambodia for assistance in coordinating this research.



  • Collaborators Not applicable.

  • Contributors TS conceptualised the study, designed the survey questionnaire, conducted the fieldwork, analysed the data and drafted the manuscript. HY, VW and JWT assisted with data analysis and provided critical revisions of the manuscript. CC assisted with coordinating data collection and provided critical revisions of the manuscript. All authors have seen and approved the final manuscript.

  • Funding This research was funded by TS’s PhD scholarship from Deakin University.

  • Competing interests None declared.

  • Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

  • Provenance and peer review Not commissioned; externally peer reviewed.