To assess the packaging of cigarettes and HTPs sold in Korea, we used a market-based purposive sampling strategy designed to capture the diversity of unique brands and pack variants available on the market, rather than to estimate sales volume or market share [17]. Our pack documentation and photography procedures were informed by the TPackSS protocol [17], and we coded pack-level flavor cues such as descriptors, imagery, and capsules based on prior pack surveillance research [11,12]. Furthermore, drawing on prior studies [30-32], we developed an exploratory protocol to cut open individual cigarettes and HTP sticks (hereafter referred to as sticks) to examine internal flavor-related features. We also documented external flavor cues, including capsule indicators on sticks, before disassembly.

The product identification process involved four steps: (1) compiling an initial product list, (2) purchasing available products, (3) cross-referencing with external data sources, and (4) investigating reasons for unavailable products (Figure 1). To construct an initial product list, we visited 1 retail store from each of the 2 major convenience store chains in Seoul (CU and 7-Eleven) in November 2024 to collect all available tobacco products. We focused exclusively on convenience stores because they represent the predominant retail channel for tobacco sales in South Korea [29]. Among these, CU and 7-Eleven were selected due to their market dominance; together, they account for 56.3% of all convenience stores nationwide [33]. Within each chain, 1 store in Seoul was selected based on geographic proximity to the research team. Observational checks at additional outlets suggested that product availability was comparable across stores.

All tobacco products available in the visited stores were purchased, including all unique retail pack variants available at the point of sale. The initial list of purchased products was cross-checked against 2 external sources: the Wave 3 (2023) International Tobacco Control Korea Survey [34], which listed 160 cigarettes and 67 HTP brands, and a government-commissioned tobacco product market report [35], which identified 225 cigarettes and 72 HTP products purchased from 46 convenience stores nationwide in October 2023.

Products missing from the initial list (142 products: 107 cigarettes and 35 HTPs) were assessed using external sources to determine their market status. Products were classified as discontinued or low-demand based on Euromonitor reports (market share <0.3%) [26] and corroborating evidence from news articles and websites describing products as “discontinued,” “low-demand,” or “unpopular.” Duty-free exclusive products were verified through media reports. Among missing cigarettes, 90 (84.1%) were discontinued, 13 (12.1%) were not stocked due to low-demand, 3 (2.8%) were out of stock at the time of purchase, and 1 (0.9%) was duty-free exclusive and therefore excluded. Among missing HTPs, 31 (88.6%) were discontinued, and 4 (11.4%) were not stocked due to low demand.

Products not classified as discontinued, low-demand, or duty-free exclusives were subsequently sought through additional visits in July 2025, resulting in the collection of 3 additional packs. In total, 254 cigarettes and 99 HTPs were identified across all sources, of which 104 (40.9%) cigarettes and 35 (35.3%) HTPs were not collected. Reasons for noncollection are documented in Tables S1 and S2 in Multimedia Appendix 1.

Descriptive analyses and data tabulation were conducted using Microsoft Excel. All characteristics were assessed according to tobacco product type (cigarettes vs HTPs). Categorical variables were presented as frequencies and percentages, while continuous variables were summarized using the minimum, median, mean, and maximum values. Continuous variables were additionally categorized, where applicable, to facilitate comparisons. Differences between cigarettes and HTPs were examined using chi-square tests in SAS (version 9.4; SAS Institute Inc), with a 2-sided significance level of P<.05. For variables with sparse expected cell counts (n<5), Fisher exact tests were used instead. For flavoring components, which allowed multiple responses, each component was treated as a binary variable (present vs absent). Proportional differences between cigarettes and HTPs were evaluated using separate chi-square tests.

This study did not involve human participants or animals. The data were collected from publicly available commercial tobacco products. Therefore, ethics approval was not required and was not sought.

A total of 214 tobacco products were included in the final analytic sample, comprising 150 cigarettes and 64 HTPs. The descriptive characteristics of the collected products are provided in Table 2 and Table S3 in Multimedia Appendix 1. More than half of cigarettes (50%) and HTPs (51.6%) were manufactured by KT&G. The average tar and nicotine contents displayed on cigarette packs were 2.70 mg (range 0.1‐8.0 mg) and 0.24 mg (range: 0.01‐0.7 mg), respectively. The display of these contents is not mandatory for HTPs, and none of the HTP packs included them. All products complied with mandated health warnings.

The main differences between cigarettes and HTPs were pack size (P<.001) and shape (P<.001) (Figure S6 in Multimedia Appendix 1). Cigarette sticks were longer, resulting in taller, vertically oriented rectangular packs with variation in depth. HTP packs were shorter, reflecting the shorter stick length, and had a consistent, landscape-oriented rectangular shape.

Although no distinct qualitative differences in promotional elements were observed between cigarettes and HTPs, a higher proportion of cigarette packs (68.7%) included marketing texts (P=.030) compared with HTPs (53.1%). Most of these statements highlighted hand and mouth odor reduction technology. For example, texts such as “activated carbon filter,” “nano capsule technology,” and “smartcore induction system” were frequently displayed, serving as marketing claims to emphasize the companies’ technological capabilities.

The distribution of flavor cues on packaging is summarized in Table 3. Among the 5 components of pack-level flavor cues, pack color was the most common, appearing on 97 (64.7%) cigarettes and 61 (95.3%) HTPs (P<.001). Within pack color cues, cool or minty shades (blue, green, and sky blue) were most frequent, followed by fruity (red, pink, and orange), nutty (brown, dark brown, and yellow), and others (dark green and purple). The second most frequent cues varied by product type, but product names and capsule images were frequently used in both. Product names often implied coolness, such as “aqua,” “cool shot,” “frozen,” and “ice.” Capsule images, typically placed on the front of the pack, signaled flavor through colors associated with specific tastes. For example, blue tones for menthol or ice and red tones for cherry or berry. Pack imagery was less frequently used but featured cool- or minty-related images such as animals (eg, penguins) or natural scenes (eg, glaciers), reinforcing coolness, and other cues included promotional terms such as “double capsule” and “refreshing scent.” In addition, capsule imagery on the stick was observed in 71 (47.3%) cigarettes and 37 (57.8%) HTPs (P=.16). Among products with capsule imagery, cigarettes were predominantly cool or minty (59.2%), whereas HTPs were more evenly distributed across cool or minty (27%), fruity (29.7%), and other categories (27%).

Figure 2 presents examples of flavor cue combinations by product. Overall, 101 (67.3%) cigarettes and 61 (95.3%) HTPs included at least 1 flavor cue (P<.001). For cigarettes, color and capsule images were commonly used together, with the most frequent combinations being “product name + color + capsule image” (11.3%; Figure 2A) and “color + capsule image” (10.7%; Figure 2B). All HTPs with flavor cues included pack color as one of the cue elements, often reinforced by additional cues such as product name and capsule imagery (Figures 2D and 2E).

Of the 214 products analyzed, 143 (66.8%) contained at least 1 flavoring component (Table 3). Flavoring was identified in 89 (59.3%) cigarettes and 54 (84.4%) HTPs (P<.001). Crushable capsules were the most common flavoring method in both cigarettes (48.7%) and HTPs (57.8%) (P=.22). In cigarettes, flavor additives were also found in cellulose filters (14.7%) and tobacco leaves (12%). In contrast, HTPs used a wider range of flavoring methods, including tobacco leaves (45.3%), cigarette wrappers (23.4%), and hollow tubes (23.4%).

Although up to 9 flavoring components (7 for cigarettes) were possible, both cigarettes and HTPs showed a maximum of 4 components per product. Overall, 25.4% of cigarettes and 54.7% of HTPs used 2 or more flavoring methods. Among products with crushable capsules, 54.8% of cigarettes and 45.9% of HTPs used capsules as the sole flavoring method. In both product types, up to 3 additional components were combined with capsules. Among capsule-containing products that used at least 1 additional flavoring component, the most common paired components were cellulose filters (51.5%) and tobacco leaves (39.4%) in cigarettes. For HTPs, tobacco leaves (70%) and inner wrappers (25%) were the most frequently paired. The distribution of paired components differed significantly between cigarettes and HTPs (P<.001), with a higher proportion of HTPs containing flavoring in tobacco leaves (70%) compared with cigarettes (39.4%).

Across all products analyzed, 63.1% (54% of cigarettes and 84.4% of HTPs) both had flavor cues on the pack and contained flavor additives (Figure 3A), while 3.7% (5.3% of cigarettes and 0% of HTPs) contained flavor additives without any packaging cues (Figure 3B). Conversely, 12.6% (13.3% of cigarettes and 10.9% of HTPs) displayed cues but lacked detectable flavor additives (Figure 3C), and 20.6% (27.3% of cigarettes and 4.7% of HTPs) showed neither cues nor additives (Figure 3D).

We conducted systematic surveillance of cigarettes and HTPs sold in Korea, integrating pack- and stick-level coding of flavor cues with component-based assessment of flavor additives. Our analysis showed that pack sizes and shapes were not standardized and varied across both product types. Pack color was the most common flavor cue, followed by capsule imagery and product names, often used in combination. Most products used 2-3 cues simultaneously, reinforcing flavor signals. Flavor cues and additives were more prevalent among HTPs than cigarettes. Although capsules were the dominant flavoring method, additives were also incorporated into other components.

The flavor cues identified align with previous tobacco pack surveillance studies documenting widespread use of packaging elements to signal flavor. However, regulatory environments shape how these cues manifest. In markets without explicit restrictions, such as the Philippines, 58.5% of tobacco products contained flavor cues through direct descriptors beneath the product name [11]. In contrast, Korea’s Article 9‐3 of the NHPA prohibits explicit flavor descriptors such as “menthol” or “cherry,” including text or imagery, on tobacco packaging [27]. Nevertheless, flavoring continued to be implied through indirect marketing strategies, using terms such as “Ice” or “Tropical” and capsule imagery on both packs and sticks. Consequently, the prevalence of flavor cues among the products collected in Korea (75.7%) was far higher than the average among countries without flavor bans (22.7%) [12]. These findings suggest that regulatory approaches limited to explicit flavor indicators may be insufficient to eliminate flavor-related marketing.

A key contribution of this study is the combined assessment of flavor cues across both packs and sticks, alongside component-based screening of flavoring additives. Over 60% of products combined flavor cues with detectable flavor additives, but cues were also present in products without detectable flavor additives. Pack-level cues were often reinforced at the point of use, as 47% of cigarettes and 58% of HTPs featured capsule imagery on the stick. This pattern highlights that flavor-related appeal can be generated independently or synergistically through perceptual mechanisms (pack and stick design) [1,47] and material product features (additives) [48]. Linking consumer-facing signals with product-level evidence, therefore, provides a more comprehensive characterization of flavored tobacco products.

These findings have practical implications for enforcing flavor ban policies. Comprehensive flavor bans alone may be insufficient, as they leave room for circumvention. Even without flavor additives, packaging elements that imply flavor can shape perceptions of taste, quality, and harm [41] and may even modulate craving-related neural responses, reinforcing product appeal [49]. Moreover, in jurisdictions where flavor bans have been implemented without plain packaging, the industry has introduced pack- and stick-level design features that insinuate flavor characteristics despite the absence of explicit additives [50]. Flavor bans and plain packaging covering both packs and sticks should therefore be implemented in tandem. However, enforcing flavor bans can be technically complex due to ambiguity in defining “characterizing flavor” [51] and limited standardized chemical testing capacity [52]. In this context, plain packaging may serve as a more immediately implementable strategy while regulatory systems strengthen enforcement capacity for flavor bans.

Several areas for future research follow from our findings. The high prevalence of flavoring additives among sampled products is broadly consistent with reported sales patterns of flavored tobacco products in Korea [26], highlighting a discrepancy between market patterns and the relatively lower self-reported use of flavored tobacco noted in Korea [53]. Only 29.8% of Korean smokers reported using flavored tobacco products (excluding capsule products), whereas only 26.7% reported that their usual brand contained a flavor capsule [26], which is lower than both the market share of flavored products and the prevalence of additives detected in our assessment. Our findings show that while crushable capsules were common, flavor additives were also frequently incorporated into less visible components such as tobacco leaves, wrappers, and internal filters. Because flavoring in these components is less noticeable and not widely recognized as a marker of flavored tobacco, such products may go unrecognized by consumers as flavored. Future research could examine whether this gap reflects underrecognition of these less salient forms of flavor delivery, or other factors such as higher consumption intensity among flavored-product users [54]. Additionally, the high proportion of discontinued products identified from our sampling procedures suggests rapid market turnover. Continued surveillance research will therefore be important to track evolving product design and flavoring strategies.

Given that evidence on HTP flavoring remains limited globally [55], our findings provide early empirical evidence in this area. Compared with cigarettes, HTPs use more pervasive cues and diverse flavoring methods, often using multiple flavoring methods across various components rather than being limited to capsules alone. As HTP markets continue to expand [56], surveillance frameworks developed for cigarettes, including pack- and stick-level assessment and evaluation of flavor additives, should be extended to HTPs. Our methods and findings offer a preliminary framework that may be applied in other settings to systematically assess the characteristics of HTPs.

Another global implication of our study is that the patterns documented in Korea may extend beyond the domestic market. KT&G, a major Korean tobacco manufacturer and global exporter [57-60], has reported increasing exports of flavored products and capsules [59], with packaging designs in some export markets that are identical or highly similar to those sold in Korea [60,61]. Surveillance of products sold in Korea may therefore provide an early signal of evolving industry tactics that could emerge in other jurisdictions.

Our study has several limitations. First, products were purchased in one metropolitan area and primarily from major convenience store chains, so availability may differ in other regions or retail channels (eg, supermarkets and duty-free outlets). Second, some identified products could not be collected due to discontinuation or low demand, which may have introduced selection bias and affected estimates of flavor cues and additives. For example, niche or experimental flavored variants are more likely to be dropped from retail shelves, potentially leading to an underestimation of the diversity or prevalence of flavor-related features. Third, coding was conducted through team-based consensus rather than independent double coding, which may have introduced bias and precluded calculation of intercoder reliability. Fourth, our sensory screening relied on a pragmatic, nonblinded orthonasal assessment conducted by trained research staff rather than professional sensory analysts. Subtle additives may have been missed, and component attribution should be interpreted as approximate due to potential odor diffusion. Future work should incorporate blinded sensory testing conducted by expert sensory panelists and chemical analyses to improve the detection and localization of flavoring additives.

To our knowledge, this study is among the first to simultaneously examine flavor cues on cigarette and HTP packaging and sticks, alongside the presence of flavor additives. By integrating pack- and stick-level assessments, we provide evidence on how flavor cues and additives work together to enhance product appeal. Our findings show that flavoring in tobacco products extends beyond visible pack design or crushable capsules to multiple structural components, highlighting the need for regulations that address both perceptual mechanisms and sources of flavoring. Incorporating stick-level monitoring and HTPs into existing surveillance systems will be crucial for identifying concealed flavoring cues and methods. These efforts can strengthen the implementation of World Health Organization (WHO) FCTC Articles 9 and 10 (regulation and disclosure of tobacco product contents and emissions) and Article 11 (packaging and labeling).