The effect of standardised packaging and limited flavour descriptors of vape pods among adults and youth in Great Britain: a cross-sectional between-subjects experimental study

Summary

Background

Vape packaging often includes colourful designs and conceptual flavour names, which appeal to youth. We examined the impact of standardised vape packaging and flavour names on appeal to youth and adults.

Methods

This cross-sectional (between-subjects) experimental study used data from the 2024 Action on Smoking and Health Youth (n = 2,770, aged 11–18), and a 2024 Prolific Academic (n = 3,947, aged 18+), surveys. Participants were randomised to one condition from: (1) branded packs (control), (2) standardised white with usual descriptors (e.g., blu razz) (3) standardised white with limited flavour descriptors (e.g., blueberry and raspberry), and among adults only (4) standardised white with coded flavour descriptors (e.g., FR248). Youth reported perceived peer interest in trying the products, adults reported their own interest and perceived harm relative to cigarettes. Logistic regression examined associations between packaging and interest or harm perceptions, adjusting for demographics.

Findings

Differences between packaging types were small in percentage terms, but some reached significance. Youth reported higher odds of no perceived peer interest in trying vapes in standardised packs (31.5% (n = 292/927); aOR = 1.85 [95%CI = 1.47–2.32], p < 0.001), and standardised packs with limited flavour descriptors (27.2% (n = 252/927); 1.50 [1.19–1.90], p < 0.001), vs. branded packs (24.0% (220/916)). Adults reported higher odds of no interest in trying standardised packs with coded flavour descriptors (76.2% (776/1019); 1.52 [1.18–1.97], p < 0.001), but not usual (75.4% (n = 713/946); 1.28 [0.99–1.65], p = 0.063) or limited (70.1% (n = 657/937); 0.90 [0.70–1.15], p = 0.392) flavour descriptors, compared to branded (71.5% (n = 693/969)). Packaging condition did not have a significant effect on relative harm perceptions among adults.

Interpretation

Standardised packaging and limiting flavour descriptors reduced youth appeal but had minimal effect on adults’ interest or harm perception.

Funding

Research England.

Research in context.

Evidence before this study

Vape (e-cigarette) packaging varies substantially across device types and brands, with many featuring vibrant colours and images to promote products. Vapes can be a useful smoking cessation tool, so therefore need to be attractive to adults who smoke whilst not appealing to youth. Standardised packaging policy, including limiting flavour descriptors, has been reported to reduce the attraction of tobacco cigarettes. Regulation to standardise vape packaging is currently being considered in multiple jurisdictions. To assess evidence on impact of potential regulations, we searched PubMed from database inception to November 6, 2024, using the following terms ((Standard∗) AND (Vape or E-cigarette)) AND (Packaging). Of the 71 results, four reported the effects of standardising vape packaging on youth and adults. Evidence suggested that standardising the packaging of vapes may reduce youth, but not adult, appeal. We did not identify any studies reporting the effects of standardising flavour names on appeal. One study reported that standardising packaging may increase misperceptions of vaping harms relative to tobacco cigarettes among youth. No studies reported on the effects on adults’ harm perceptions.

Added value of this study

This study found that standardising packaging and limiting flavour descriptors reduced youth’s, but not adults’, interest in trying vapes. There was little effect of standardised packaging on adults’ relative harm perceptions.

Implications of all the available evidence

In 2024, the United Kingdom Government announced intentions to introduce powers to regulate vape packaging and flavours. Vape packaging and flavour regulations need to strike a delicate balance between not dissuading people who smoke from using the products to quit smoking, whilst also dissuading youth and people who have never smoked from trying them. Our findings in combination with previous research suggest that a policy to standardise packaging in combination with standardised or alphanumeric flavour descriptors could reduce youth and non-smoking adults’ interest in trying vaping products, without dissuading adults who smoke from trying them or inflating misperceptions of harm.

Introduction

There is robust evidence that vapes (e-cigarettes) present significantly lower levels of harm compared to tobacco cigarettes, which kill one in two long term users.^1,2 Vapes have also been shown to successfully help adults who smoke to quit.³ However, vapes are not risk free and therefore should only be marketed towards adults who smoke and should not be attractive to youth.

Packaging is a primary marketing tool for vape companies. Vape packaging varies substantially across device types and brands, with many brands featuring vibrant colours and images to promote products.⁴ From being visually appealing in shops (point of sale), to featuring in online advertisements and social media campaigns, packaging is key to brand identity and promotion.⁵ As well as bright colours, it is commonplace for brands to also use youth-appealing elements, such as images, cartoon characters, stylised fonts, and novel brand and flavour names on vapes and e-liquid packaging.⁴ The packaging of vapes influences the appeal of vaping products to youth and young adults,⁶ and both youth and adults report that reducing bright colours and branding features would make vapes less appealing to youth.⁷

Flavours play a key role in both youth and adult initiation and continuation of vaping,⁸ with fruit flavours being the most popular among these groups in GB.⁹ Vape and e-liquid packaging often use images, sensory descriptors, and conceptual names (e.g., ‘blue voltage’) to describe flavours,^4,10 which are popular among youth.¹¹ Flavour-related packaging elements have been found to influence perceptions among youth aged 11–14 in the US, with packaging with fruit-related colours and imagery creating the strongest perceptions of novelty (e.g., ‘fun’, ‘interesting’).¹²

In 2024, in an effort to reduce the appeal of vapes to youth, the United Kingdom (UK) government proposed powers to regulate the marketing, packaging, and flavours of nicotine-containing products including vapes. However, the exact details and regulatory timeline of these powers have yet to be determined.¹³ Specific regulations have not yet been announced, but standardising colours, removing branding on packs, regulating flavour name descriptors, and replacing flavour and brand names with alphanumeric codes are all potential options. Previous US policy has often focused on restricting flavours, but neglect whether change to much more visible packaging could achieve desired effects. Israel and Denmark already require standardised vape packaging,^14,15 the Netherlands required standardised packaging from July 2025,¹⁶ and flavour names are regulated in New Zealand¹⁷; however, the effects of these polices have not yet been evaluated. The UK and many other jurisdictions have already introduced similar regulations for tobacco cigarettes. Bans on flavour and brand descriptions, and requiring packs to be sold in a standardised olive green has been shown to reduce the appeal of tobacco cigarettes to youth and adults.^18,19

The government also announced a ban on disposable vaping products under the 2024 Environmental Protection Act.¹³ In response, many vape brands have shifted their popular disposable varieties to vape pods; these are pre-filled e-liquid cartridges designed to be used with reusable pod vape devices, and hence not affected by the disposable ban.

Contrary to tobacco cigarette packaging, vape packaging regulations need to strike a delicate balance between not dissuading people who smoke from using the products to quit smoking, whilst also dissuading youth and people who have never smoked from trying them. Our previous research has found significant reductions in interest in trying vape starter kits when packaging is standardised among youth but not adults,⁶ and that youth are less interested in trying e-liquids when they are in standardised packaging with coded flavour and brand names, compared to branded packs.^6,20,21 Research among cigarette has also shown that alphanumeric coded brand names can reduce appeal,²² however this has not yet been researched for flavour names.

Standardised packaging has been found to increase the inaccurate perception that vaping is equally/more harmful than smoking among youth.²¹ This is problematic because inaccurate vaping harm perceptions can deter adults who smoke from switching to a less harmful product.¹ To our knowledge, there is no current research on the effects of limiting flavour names on interest in trying vapes or perceptions of their harms among adults.

Therefore, we aimed to examine associations between 1) standardised vape pod packaging, limited flavour descriptors, and interest in trying the vape pods displayed, among adults and youth (and whether associations differed by smoking and vaping status); and 2) standardised vape pod packaging and vape pod harm perceptions relative to cigarettes, among adults only.

Methods

Study design

In this between-subject experimental design, data were collected through online experiments embedded into two surveys: 1) the Action on Smoking and Health 2024 Smokefree Great Britain Youth Online Survey (ASH-Youth); 2) an online survey among adults run specifically for this study.

Youth data provided were obtained from the 2024 Smokefree Great Britain Youth Online Survey collected between 21 March and 18 April. Respondents were drawn from an existing online panel maintained by YouGov. Active sampling was used, which dynamically evaluates what surveys are available for a particular panel member based on their personal characteristics, and restrictions are put in place to ensure that only those who are selected from their panel of registered users are allowed to take part in the survey. Informed consent was provided either by the parents of those aged 11–15 years or by individuals aged 16–18 years.

Adult data were obtained from an online survey, collected between 24 May and 4 June 2024 using Prolific Academic, a participant recruitment community for science. Respondents were from a pre-existing panel with quota samples set to ensure the sample was representative of age, sex, and ethnicity in the UK. All respondents provided informed consent.

The protocol was pre-registered on the Open Science Framework https://osf.io/849nk/.

Procedures

The surveys included a cross-sectional between-subject experimental design where, using simple randomisation, youth respondents were randomly allocated at a 1:1:1 ratio to one of three conditions (1–3), and adults at a 1:1:1:1 ratio to one of four conditions (1–4). Within each condition, respondents viewed a set of the same four brands of vape pods in a range of fruit flavours. The conditions included images of vape pods in:

• 1.Fully branded packaging (control),
• 2.Standardised packaging with usual flavour descriptors,
• 3.Standardised packaging with limited flavour descriptors,
• 4.Standardised packaging with coded flavour descriptors (adults only).

The four brands were chosen to represent a range of popular UK brands and the most popular flavour category (fruit) in the UK.²³ Packaging was standardised in white, as opposed to the green/brown Pantone 448C used for standardised tobacco cigarette packaging in the United Kingdom, to ensure a distinction between vapes and tobacco cigarettes, because of differences in relative harm and consistent with prior work in this area.²⁰ Due to space limitations within the ASH-Y survey, it was not possible to randomise youth to the standardised packaging with coded flavour descriptors; therefore, this condition was only present in the adult survey. Participants were randomised automatically by the online survey software. They only viewed one condition and were not aware of other conditions.

Participants

The ASH-Youth survey was completed by N = 2872 youth aged 11–18. Respondents who reported “Prefer not to say” or had missing data for the outcome (n = 20) or any covariates (n = 82) were excluded from the sample, resulting in a final analytic sample of N = 2770 respondents.

The Prolific Adult survey was completed by N = 4040 adults aged 18+. Respondents who reported “Prefer not to say”, “Don’t know” or had missing data for any covariates (n = 93) were excluded from the sample, resulting in a final analytic sample of N = 3947 respondents.

Measures

Youth

Details on coding are reported in Supplementary file 1.

Outcome measure

Primary: No perceived peer interest in trying

Respondents were shown a set of images of vape pods from four different brands based on the experimental condition they had been randomised to (Fig. 1) alongside a question that asked, “Which of the following products would people your age be most interested in trying?”. Response options included choosing one of the four vape pod images, or selecting ‘None of these products,’ ‘I don’t know,’ or ‘Prefer not to say’. There was no time limit on the question. Responses were coded ‘None of these products’ vs. any other response. Respondents who selected “Prefer not to say” (n = 20) were excluded.

Fig. 1.

Packaging conditions.

Questions were worded differently for youth and adults (see below) because of ethical concerns about asking youth—particularly those who had never vaped—whether they would be interested in trying a product that they may not have been exposed to prior to this study. Also, due to space limitations within the survey, youth were not asked about their harm perceptions of products.

Vaping/smoking measures

Vaping status

‘Currently vaping, ‘Past/tried vaping’, and ‘Never vaped/never heard of vapes’. Don’t know responses (n = 20) were excluded.

Smoking status

‘Currently smoking’, ‘Past/tried smoking’, ‘Never smoked’. ‘Don’t know’ responses (n = 59) were excluded.

Demographic measures

Age, Sex, Socioeconomic status (SES). Missing SES responses (n = 13) were excluded. See Supplementary file 1 for detail. Demographic covariates have previously been reported to be associated with vaping prevalence.⁹

Adult

Full questionnaires and data analysis pre-registration are available on the Open Science Framework (https://osf.io/849nk/) and details on coding are reported in Supplementary file 1.

Outcome measures

Primary: No interest in trying

Respondents were shown a set of 4 images of vape pod packs based on the experimental condition they had been randomised to (Fig. 1) alongside a question (again without time limit) that asked, “Which of the following products would you be most interested in trying?”. Response options included choosing one of the four images, ‘None of these products,’ or ‘I don’t know’. Responses were coded as ‘No interest (None of these products)’ vs. ‘Interest (choosing any of the four products)’. As stated in the pre-registration, as <5% of responses were ‘don’t know’ (n = 76, 1.9%), they were excluded from analyses.

Secondary: relative harm perceptions

In the adult survey only, respondents were then randomised to just one of the four images from the condition they had already seen, and asked, “How harmful do you think it is to vape this product?”. Response options include: 1) ‘Not at all harmful’, 2) ‘Harmful, but less harmful than smoking cigarettes’, 3) ‘As harmful as smoking cigarettes’, 4) ‘More harmful than smoking cigarettes’, 5) ‘Don’t Know’. Responses were coded ‘Harmful, but less harmful than smoking cigarettes (2)’ vs. ‘Other (1, 3–5)’, consistent with current evidence about the harms of vaping.¹

Vaping/smoking measures

See Supplementary file 1 for detail.

Vaping measure

‘Currently vape’, ‘Used to vape’, ‘Never vaped’.

Smoking measure

‘Currently smoke’, ‘Used to smoke’, ‘Never smoked’.

Demographic measures

Age, Gender, Race/ethnicity, Perceived Financial Status. Gender was coded ‘male’, ‘female’, ‘in another way’. Due to small sample sizes for regression analyses, gender was coded ‘female’ and ‘other (male or in another way)’. Race/ethnicity was coded: 1) Asian, Asian British, 2) Black, Black British, Caribbean or African, 3) Mixed or Multiple ethnic groups, 4) White, and 5) Other ethnic group, in line with the Census definitions (Office for National Statistics, 2023). Due to small sample sizes for regression analysis, ethnicity was coded ‘White’ (4), and ‘Racialised minorities (1–3, 5)’. Prefer not to say responses were removed for Gender (n = 12) and Ethnicity (n = 19). For Perceived Financial Status ‘Prefer not to say’ responses (n = 61) and ‘Don’t know’ (n = 15) were removed. Demographic covariates have previously been reported to be associated with vaping prevalence.²⁴

Data analysis

Youth

To investigate perceived peer interest in trying, multinomial regressions examined associations between ‘Interest (ref)’, ‘No interest’ and ‘Don’t know’ in trying the product and vape packaging condition (branded packaging as reference). Analysis was first unadjusted, then adjusted for smoking status, vaping status, age, sex and SES. Models were then repeated, stratified by vaping and smoking status.

Adult

To investigate interest in trying, logistic regression examined associations between ‘No interest’ in trying the product and vape packaging condition (branded packaging as reference). Analysis was first unadjusted, and then adjusted for smoking status, vaping status, age, gender, ethnicity and PFS. Models were then repeated, stratified by vaping and smoking status.

To investigate harm perceptions, logistic regression examined associations between relative harm perceptions and vape packaging condition (branded packaging as reference). Analysis was first unadjusted, and then adjusted for smoking status, vaping status, age, gender, ethnicity and PFS. Models were then repeated, stratified by vaping and smoking status.

Deviation from the pre-registration

Due to a large proportion of ‘Don’t Know’ responses among youth (N = 788, 28.4%) for questions on perceived peer interest, multinomial models were utilised instead of pre-stated binomial models so ‘Don’t Know’ could be considered its own category. Due to a small proportion of ‘Don’t Know’ responses among adults (n = 76, 1.9%) for questions on interest, planned sensitivity analyses comparing interest, no interest and ‘Don’t Know’ were not conducted.

Due to small cell counts for youth who currently smoked, differences in outcomes by smoking and vaping status were analysed by stratifying regressions by smoking and vaping status instead of introducing interaction terms.

Ethical approval

For youth data, ethical approval for the analyses in this study was not required because this study involved secondary analysis of preexisting data, in line with King’s College London policy. For adult data collection, ethical approval was granted by King’s College London (MRA-23/24-42484).

Role of the funding source

The funder of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report.

Results

Youth

There were similar proportions of male (51.4%) and female participants (48.6%), with most from a high socioeconomic status (ABC1 70.0%, C2DE 30.0%; see Supplementary file 1 for definitions). Few participants currently vaped (8.4%) or smoked (5.8%) (Supplementary file 2, Table 1). Among youth who vaped, most used fruit flavours (65.4%) and disposable devices (55.8%) (Supplementary file 2, Table 2).

Table 1.

Multinomial associations between perceived peer interest in trying vape pods displayed, packaging condition and participant characteristics; ASH-Youth 2024 (n = 2770).^a

	Interest in trying any pod displayed (ref)	No interest in trying any pod displayed			Don’t know
	% (n)	% (n)	AOR (95%CI)	p	% (n)	AOR (95%CI)	p
Total	43.9 (1217)	27.6 (765)			28.4 (788)
Packaging condition
Branded packaging	53.4 (489)	24.0 (220)	1	ref	22.6 (207)	1	ref
Standardised packaging with usual flavour descriptors	38.2 (354)	31.5 (292)	1.85 (1.47–2.32)	<0.001	30.3 (281)	1.91 (1.52–2.40)	<0.001
Standardised packaging with limited flavour descriptors	40.4 (374)	27.2 (252)	1.50 (1.19–1.90)	<0.001	32.4 (300)	1.91 (1.52–2.42)	<0.001
Gender
Male	37.9 (315)	26.9 (224)	1	ref	35.2 (293)	1	ref
Female	47.3 (636)	24.9 (335)	0.71 (0.59–0.86)	<0.001	27.8 (274)	0.83 (0.69–1.00)	0.050
Age group
11–15	40.9 (709)	30.6 (530)	1	ref	49.4 (285)	1	ref
16–18	49.0 (509)	22.6 (235)	0.77 (0.63–0.94)	0.009	28.3 (294)	0.99 (0.82–1.21)	0.968
SES
ABC1	46.6 (903)	27.9 (540)	1	ref	25.5 (495)	1	ref
C2DE	37.9 (315)	26.9 (224)	1.18 (0.96–1.46)	0.121	35.2 (293)	1.69 (1.38–2.06)	<0.001
Vaping status
Never vapedb	39.6 (880)	31.4 (698)	1	ref	30.1 (669)	1	ref
Past/tried vaping	58.1 (226)	13.5 (43)	0.29 (0.20–0.43)	<0.001	24.8 (79)	0.64 (0.47–0.87)	0.004
Currently vaping	69.4 (111)	10.4 (24)	0.25 (0.15–0.40)	<0.001	17.3 (40)	0.43 (0.28–0.66)	<0.001
Smoking status
Never smoked	39.6 (880)	29.8 (662)	1	ref	30.6 (679)	1	ref
Past/tried smoking	58.1 (226)	21.6 (84)	0.89 (0.65–1.21)	0.442	20.3 (79)	0.60 (0.44–0.82)	0.001
Currently smoking	68.4 (111)	11.9 (19)	0.53 (0.30–0.93)	0.026	18.8 (30)	0.59 (0.37–0.96)	0.033

p < 0.05 are indicated in bold.

^aModel is adjusted for gender, age, SES, vaping status and smoking status.

^bNever vaped includes those who have never heard of vaping.

Table 2.

Logistic regression associations between interest in trying vaping pods, packaging condition and participant characteristics; adults in the UK 2024 (N = 3871).

	Interest in trying any pod displayed (ref)	No interest in trying any pod displayeda
	% (n)	% (n)	AOR (95%CI)	p
Total	26.7 (1032)	73.3 (1548)
Packaging condition
Branded packaging	28.5 (276)	71.5 (693)	1	ref
Standardised packaging with usual flavour descriptors	24.6 (233)	75.4 (713)	1.28 (0.99–1.65)	0.063
Standardised packaging with limited flavour descriptors	29.9 (280)	70.1 (657)	0.90 (0.70–1.15)	0.392
Standardised packaging with coded flavour descriptors	23.8 (243)	76.2 (776)	1.52 (1.18–1.97)	0.001
Gender
Male and other	28.7 (537)	71.3 (1331)	1	ref
Female	24.7 (495)	75.3 (1508)	0.89 (0.74–1.06)	0.192
Ethnicity
White	25.3 (847)	74.7 (2496)	1	ref
Racialised minorities	35.0 (185)	65.0 (343)	0.72 (0.56–0.93)	0.013
Age
18–24	37.5 (152)	62.5 (253)	1	ref
25–34	41.6 (281)	58.4 (394)	0.76 (0.55–1.04)	0.087
35–44	33.6 (215)	66.4 (424)	0.98 (0.70–1.37)	0.909
45–54	26.5 (176)	73.5 (487)	1.61 (1.14–2.28)	0.006
55 or older	14.0 (208)	86.0 (1281)	2.64 (1.90–3.66)	<0.001
SES
Comfortable	20.9 (227)	79.1 (860)	1	ref
Coping	25.9 (479)	74.1 (1372)	0.96 (0.77–1.21)	0.733
Finding it difficult	34.2 (230)	65.8 (443)	0.88 (0.66–1.16)	0.349
Finding it very difficult	36.9 (96)	63.1 (164)	0.78 (0.54–1.14)	0.198
Vaping status
Currently vape	71.9 (485)	28.1 (190)	1	ref
Used to vape	41.5 (405)	58.5 (570)	3.24 (2.58–4.07)	<0.001
Never vaped	6.4 (142)	93.6 (2079)	18.59 (13.92–24.84)	<0.001
Smoking status
Currently smoke	62.7 (444)	37.3 (264)	1	ref
Used to smoke	31.2 (381)	68.8 (841)	1.97 (1.57–2.47)	<0.001
Never smoked	10.7 (207)	89.3 (1734)	3.55 (2.68–4.71)	<0.001

p < 0.05 are indicated in bold.

^aModels are adjusted for vaping status, smoking status, age, gender, ethnicity and perceived financial status. Due to small cell counts, ethnicity was collapsed into ‘White’ and ‘Racialised minorities’. Gender was also collapsed into ‘Female’ and ‘Other’. Don’t know responses (N = 76, 1.9%) were removed from analysis, in line with the pre-registration (if <5% don’t know this response is excluded from analyses).

Perceived peer interest in trying

In unadjusted and adjusted analyses, youth had significantly greater odd of reporting that they believed peers would not be interested in trying products in standardised packs with usual flavour descriptors (31.5% [n = 292/N = 927], aOR = 1.85, 95% CI = 1.47–2.32; p < 0.001), and standardised packs with limited flavour descriptors (27.2% [n = 252/N = 927], aOR = 1.50, 95% CI = 1.19–1.90, p < 0.001), compared to branded packs (24.0% [n = 220/N = 916]).

Youth had significantly greater odds of reporting that they did not know if their peers would want to try products in standardised packaging with usual flavour descriptors (30.3% [n = 281/N = 927], aOR = 1.91, 95 CI = 1.52–2.40, p < 0.001), and standardised packaging with limited flavour descriptors (32.4% [n = 300/N = 927], aOR = 1.91, 95% CI = 1.52–2.42, p < 0.001) compared to branded packs (22.6% [n = 207/N = 916]) (Table 1, Fig. 2a).

Fig. 2.

a: Youth perceived peer interest in trying vape pods by packaging condition; ASH-Youth 2024 (n = 2770). b: Adult interest in trying vape pods by packaging condition; Prolific Academic 2024 (n = 3871). For a and b, ∗ indicate significant comparisons (p < 0.05).

When standardised packaging with usual flavour descriptors was compared to standardised packaging with limited flavour descriptors, there were no significant differences in odds of perceiving that peers would not be interested in trying products (31.5% vs. 27.2%; aOR = 1.23, 95% CI = 0.98–1.55, p = 0.080) or ‘don’t know’ responses (30.3% vs. 32.4%; aOR = 0.98, 95% CI = 0.80–1.25, p = 0.983).

Perceived peer interest in trying by smoking status

When analyses were stratified by smoking status, youth who had tried smoking in the past had significantly greater odds of reporting that peers would not be interested trying products displayed in standardised packaging with usual flavour descriptors (25.2% [n = 37/N = 147], aOR = 3.04, 95% CI = 1.47–6.27, p = 0.003), and standardised packaging with limited flavour descriptors (25.8% [n = 33/N = 128], aOR = 2.31,955 CI = 1.12–4.80, p = 0.024) compared to branded packs (13.0% [n = 15/N = 115]). Similarly, youth who had never smoked had significantly greater odds of reporting that peers would not be interested in trying products in standardised packaging with usual flavour descriptors (33.6% [n = 248/N = 739], aOR = 1.81, 95% CI = 1.41–2.32, p < 0.001), and standardised packaging with limited descriptors (29.5% [n = 218/N = 739], aOR = 1.54, 95% CI = 1.30–1.99, p < 0.001) compared to branded packs (26.2% [n = 195/N = 743]). There was little evidence for a difference in perceived peer interest between packaging conditions among youth who currently smoked (Supplementary Table S3a), however, sample sizes were small (N = 19 for no interest), so findings should be interpreted with caution.

Perceived peer interest in trying by vaping status

When analyses were stratified by vaping status, youth who currently vaped had significantly greater odds of reporting that peers would not be interested in trying products displayed in standardised packaging with usual flavour descriptors (18.8% [n = 13/N = 69], aOR = 3.83, 95% CI = 1.18–12.44, p = 0.026), but not standardised packaging with limited flavour descriptors (8.0% [n = 7/N = 87]) compared to branded packs (5.4% [n = 4/N = 75]). However, sample sizes were small (N = 24 for no interest), so findings should be interpreted with caution (Supplementary Table S3b). Among youth who had never vaped, respondents had significantly greater odds of reporting that peers would not be interested in trying products in standardised packaging with usual flavour descriptors (35.3% [n = 264/N = 748], aOR = 1.93, 95% CI = 1.50–2.47, p < 0.001), and standardised packaging with limited flavour descriptors (31.1% [n = 232/746], aOR = 1.55, 95% CI = 1.21–1.99, p < 0.001) compared to branded packs (27.8% [n = 202/N = 726]). There was little significant difference in reporting don’t know between packaging conditions among youth who had tried vaping in the past (Supplementary Table S3b).

Adult

Participant characteristics are outlined in Supplementary file 2, Table S4. There were similar proportions of male (47.8%) and female (51.7%) participants, with few identifying in another way (0.5%). Most participants perceived their financial status to be comfortable (28.0%) or coping (47.8%). Under a fifth of participants currently vaped (17.7%) or currently smoked (18.8%). Among adults who vaped, most used fruit flavours (63.4%). Two fifths used tank devices (41.2%), 34.7% used disposable devices, and 23.1% pod devices (Supplementary file 2, Table S2).

Interest in trying

Overall, in unadjusted and adjusted analyses, adults had significantly greater odds of reporting that they would not be interested in trying vaping pods in standardised packaging with coded flavour descriptors (76.2% [n = 776/N = 1019], aOR = 1.52, 95% CI = 1.18–1.97, p = 0.001), compared to fully branded packs (71.5% [n = 693/N = 969). There was no significant difference in reporting no interest in trying branded packaging (71.5%) when compared to standardised packaging with usual flavour descriptors (75.4% [n = 713/N = 946], aOR = 1.28, 95% CI = 0.99–1.65, p = 0.063) or limited flavour descriptors (70.1% [n = 657/N = 937], aOR = 0.90, 95% CI = 0.70–1.15, p = 0.392) (Table 2, Fig. 2b). When standardised packs were compared, adults had significantly greater odds of reporting no interest in trying vaping products in standardised packaging with usual descriptors (75.4%; aOR = 1.43, 95%CI = 1.10–1.85, p = 0.007), or coded descriptors (76.2%; aOR = 1.70, 95%CI = 1.31–2.20, p < 0.001), compared to standardised packaging with limited flavour descriptors (70.1%). There were no other significant comparisons (Table 2, Fig. 2b).

Interest in trying by smoking status

When analyses were stratified by smoking status, adults who currently smoked had significantly greater odds of reporting that they would not be interested in trying vaping products in standardised packaging with usual flavour descriptors (44.9% [n = 75/N = 167], aOR = 1.59, 95% CI = 1.03–2.46, p = 0.037) compared to branded packs (33.9% [n = 60/N = 177]). Adults who had never smoked had significantly greater odds of reporting no interest in trying products in standardised packaging with coded flavour descriptors (93.5% [n = 473/N = 318], aOR = 2.00, 95% CI = 1.28–3.12, p = 0.002) compared to branded packs (87.8% [n = 423/N = 506]). There were no other significant differences (Supplementary Table S5).

Interest in trying by vaping status

When analyses were stratified by vaping status, adults who currently vaped had significantly greater odds of reporting that they would not be interested in trying vaping products with standardised packaging with usual flavour descriptors (34.3% [n = 54/N = 157], aOR = 1.64, 95% CI = 1.01–2.66, p = 0.046) compared to fully branded packs (24.2% [n = 40/N = 165]). Adults who had never vaped had significantly greater odds of reporting no interest in trying products in standardised packaging with coded flavour descriptors (96.6% [n = 571/N = 591], aOR = 2.30, 95% CI = 1.33–3.98, p = 0.003) compared to branded packs (92.5% [n = 509/N = 550]). There were no other significant differences (Supplementary Table S5).

Harm perceptions

Vape pods were perceived as harmful, but less harmful than smoking cigarettes by 52.8% of respondents who viewed branded packs, 49.3% for standardised packs with usual flavour descriptors, 50.5% for standardised packs with limited flavour descriptors and 51.8% for standardised packs with coded flavour descriptors (Supplementary Fig. S1). In unadjusted and adjusted analyses, there was little evidence for an effect of standardising packaging on relative harm perceptions of vaping products compared with smoking (Table 3). When standardised packs were compared, there were no significant differences in relative harm perceptions between the three types of standardised packs (p > 0.05).

Table 3.

Logistic regression associations between relative harm perception of vaping pods, packaging condition and participant characteristics; Adults in the UK 2024 (N = 3947).^a

	Product is harmful, but less harmful than smoking cigarettes (ref)	Other perceptionb
	%(n)	%(n)	AOR (95%CI)	p
Total	51.1 (2018)	48.9 (1929)
Packaging condition
Branded packaging	52.8 (525)	47.2 (469)	1	ref
White standardised packaging	49.3 (471)	50.7 (485)	1.15 (0.95–1.39)	0.141
White standardised with limited flavour descriptors	50.5 (481)	49.5 (471)	1.11 (0.92–1.34)	0.269
White standardised with coded flavour descriptors	51.8 (541)	48.2 (504)	1.05 (0.87–1.26)	0.610
Ethnicity
White	52.4 (1786)	47.6 (1625)	1	ref
Racialised minorities	43.3 (232)	56.7 (304)	1.41 (1.15–1.73)	<0.001
Age
18–24	48.3 (197)	51.7 (408)	1	ref
25–34	50.9 (349)	49.1 (685)	0.86 (0.67–1.12)	0.864
35–44	52.5 (340)	47.5 (648)	0.76 (0.58–0.99)	0.759
45–54	55.9 (377)	44.1 (674)	0.66 (0.51–0.87)	0.661
55 or older	49.3 (755)	50.7 (1532)	0.81 (0.63–1.04)	0.807
Gender
Other	60.1 (1145)	39.9 (1950)	1	ref
Female	42.8 (873)	57.2 (2042)	1.93 (1.69–2.20)	<0.001
Perceived financial status
Comfortable	51.0 (564)	49.0 (542)	1	ref
Coping	51.6 (973)	48.4 (914)	1.02 (0.87–1.19)	0.837
Finding it difficult	50.9 (351)	49.1 (339)	1.11 (0.91–1.37)	0.308
Finding it very difficult	49.2 (130)	50.8 (134)	1.25 (0.94–1.68)	0.127
Vaping status
Currently vape	78.2 (545)	21.8 (697)	1	ref
Used to vape	52.0 (519)	48.0 (999)	3.51 (2.79–4.41)	<0.001
Never vaped	42.4 (954)	57.6 (2251)	5.22 (4.07–6.69)	<0.001
Smoking status
Currently smoke	60.8 (451)	39.2 (742)	1	ref
Used to smoke	58.8 (734)	41.2 (1248)	0.73 (0.59–0.90)	0.003
Never smoked	42.6 (833)	57.4 (1957)	0.86 (0.68–1.08)	0.193

p < 0.05 are indicated in bold.

^aModels are adjusted for vaping status, smoking status, age, gender, ethnicity and perceived financial status.

^bOther perceptions’ include: ‘not at all harmful’, ‘as harmful as smoking cigarettes’, ‘more harmful than smoking cigarettes’, ‘don’t know’. Due to small cell counts, ethnicity was collapsed into ‘White’ and ‘Racialised minorities’. Gender was also collapsed into ‘Female’ and ‘Other’.

Harm perceptions by smoking status

Adults who had smoked in the past had significantly greater odds of reporting inaccurate relative harm perceptions when shown standardised packaging with coded flavour descriptors (46.5% [n = 152/N = 327], aOR = 1.57, 95% CI = 1.12–2.20, p = 0.009) compared to fully branded packs (36.8% [n = 117/N = 318]). There were no other significant differences (Supplementary Table S6).

Harm perceptions by vaping status

There were no significant differences in harm perceptions by vaping status (Supplementary Table S6).

Discussion

Among youth, especially youth who had never smoked or vaped, perceiving that peers would be not be interested in trying products increased when vaping products were in standardised packs with usual flavour descriptors or limited flavour descriptors, compared to branded packs. Adults who currently vaped or smoked were less interested in trying vapes in standardising packaging with usual flavour descriptors compared to branded packaging. Adults who had never vaped or smoked were less interested in trying coded flavour descriptors compared with branded packs. For all adults, there was no significant difference in interest in trying products in branded packaging compared to in standardised packs with limited flavour descriptors. The different packaging conditions were not associated with relative harm perceptions, except for among adults who used to smoke, where standardised packs with coded flavour descriptors increased misperceptions. Among adults who did not smoke and/or vape, there was notable disinterest in trying any of the vaping products.

Our findings on standardising packaging are broadly in line with previous research on vaping devices^6,25 e-liquids^20,21 and nicotine pouches,²⁶ among adults and youth. There was little additional effect of standardising flavour names. Interestingly, the only condition that reduced interest among adults who smoked and vaped was standardised packaging with usual flavour descriptors. Alphanumeric flavour descriptors have been previously reported to reduce youth appeal of e-liquids,²⁰ with our research finding similar effects only among adults who do not vape or smoke. Previous research has reported an impact of standardised packaging on relative harm perceptions among youth,²² however promisingly we did not find this among adults. Among youth, don’t know responses were high (28.4%), which may reflect uncertainty around the wording of the question about perceived peer interest rather than own interest. Don’t know responses were also seen to increase for both standardised packaging conditions, suggesting that removing some branding elements and descriptors make youth less able to distinguish between, and choose, a product that they believe peers would be interested in.

Overall, our findings suggest that a policy to standardise packaging could reduce youth and non-smoking adults’ interest in trying vaping products, without dissuading adults who smoke from trying them or inflating misperceptions of harm. The evidence for standardising flavour descriptors is less clear, although this should still be considered as previous evidence has highlighted the importance of flavours to youth.⁸ Alphanumeric flavour codes showed an effect on interest for adults, but may present complications in the real world, as previous research has found that adults believe that replacing cigarette brand names with alphanumeric codes would cause confusion within a retail environment.²⁷ If packaging regulations are introduced, exploitation of loopholes is possible, as was seen with the introduction of short-fill e-liquids in response to caps on e-liquid concentration and bottle size.^28,29

Some limitations and strengths need to be considered. First, there were differences between the youth and adult surveys in the wording of the items assessing interest. The ASH Youth Survey measure asked respondents about interest among people their age. However, a study among youth that asked respondents about their own interest in trying found similar results,²² suggesting peer interest is an indicator of respondents’ own interest. Second, prefer not to say responses were removed for gender, ethnicity, and perceived financial status, which may reduce generalisability to people who did not wish to disclose this information; however, overall numbers of these participants were small. Third, although the adult sample was weighted to be nationally representative of age and ethnicity, current vaping was higher than estimates from surveys that use similar measures but more thorough sampling techniques.²⁴ As our analysis shows that current vaping is significantly associated with interest in products, it is likely that the interest in product use among our sample is not representative of the UK population. Fourth, although we would not expect youth reporting of socioeconomic status to directly align with adult, in comparison to the 2021 England and Wales Census, our sample seems to be skewed towards youth of higher socio-economic status.³⁰ Fifth, sample sizes for some subgroup analyses, such as youth who smoke, were small, so findings should be interpreted with caution. However, this is the first research to explore a range of standardised packaging options and presents important policy relevant findings.

Conclusions

Standardising vape pod packaging and limiting their flavour descriptors reduces the appeal of products to youth but not adults, with little effect on adults’ relative harm perceptions. Overall, findings support policies to regulate vape packaging.

Contributors

Eve Taylor: Conceptualization, Funding acquisition, Methodology, Formal analysis, Writing—original draft. Madeleine Ebdon: Formal analysis, Visualization, Writing—original draft. Matilda Nottage: Methodology, Writing—review & editing. Erikas Simonavicius: Methodology, Writing—review & editing. Leonie Brose: Methodology, Writing—review & editing. Ann McNeill: Conceptualization, Funding acquisition, Writing—review & editing. Deborah Arnott: Conceptualization, Funding acquisition, Writing—review & editing. Hazel Cheeseman: Conceptualization, Funding acquisition, Writing—review & editing. Laura Bunce: Conceptualization, Funding acquisition, Writing—review & editing. Katherine East: Conceptualization, Funding acquisition, Methodology, Supervision, Writing—review & editing.

Data sharing statement

Data is available on reasonable request to eve.taylor@ucl.ac.uk.

Declaration of interests

The work of Action on Smoking and Health is funded by the British Heart Foundation and Cancer Research UK. KE is the recipient of Fellowship funding from the SSA and also receives salary support the US National Institutes of Health (NIH, 1P01CA200512). ET and AmN receive funding from the National Institute for Health and Care Research Health Protection Research Unit in Environmental Exposures and Health, a partnership between the UK Health Security Agency and Imperial College London. MN receives funding from CRUK (PPRCTAGPJT∖100008) and the ESRC through the London Interdisciplinary Social Science Doctoral Training Partnership. ES and LB received funding from the National Institute of Health Research Policy Research Unit in Addictions (NIHR206123). All authors declare no financial links with tobacco companies, e-cigarette manufacturers, or their representatives.