Cigarette-Dominant Users More Tobacco Dependent Than Smokeless Product, Poly Users

Cigarette-dominant tobacco users may have a higher level of tobacco dependence (TD) compared with smokeless product users and poly users who switch between cigarettes, e-cigarettes, and cigars, according to a study published in Nicotine & Tobacco Research.

Due to the increasing number of tobacco varieties, the researchers noted that it has become harder to understand tobacco use (TU) profiles’ associations with TD. Consequently, this study “aimed to identify TU profiles and their associations with TD over time, and to identify subgroups with high risk of TD.”

The study population included 3463 individuals who participated in waves 1-4 of the Population Assessment of Tobacco and Health (PATH) study and had complete TU and TD data available. The PATH study spanned from mid-September 2013 to mid-January 2018, and participants had to have been recent tobacco users—meaning current or within the last 30 days. The researchers considered 8 main tobacco product groups, including cigarettes, cigars, and smokeless.

From the data, investigators identified 3 distinct TU profiles that remained consistent across the 4 survey waves. These profiles were dominant cigarette users (62%-68%), poly users who often use traditional cigarettes, cigars, and e-cigarettes (24%-31%), and dominant smokeless product users (7%-9%).

A composite TD index and a TD sum were established based on 16 assessment items informed by prior studies. These items covered various domains, including craving, loss of control, tolerance, and withdrawal, to adequately characterize TD across a range of tobacco products. The TD sum was the simple sum of the 16 assessment items, which included 15 symptoms with a 5-point Likert scale and 1 symptom with a binary response. The TD sum ranged from 16 to 77, with higher values indicating higher TD.

The researchers used the TD-index to connect participants’ TD levels to their survey item responses via an item-response-theory (IRT) approach that connects the underlying traits being assessed—such as latent TD level—with manifestations in the form of observed responses to items on the survey. "In other words, the IRT approach models the relationship between individuals’ underlying traits and how they respond to items on a scale,” the authors explained. “Based on participants’ response to a set of items, the IRT model predicts the trait score, a value that identifies an individual’s trait level along the scale.”

The data showed that TD was significantly lower among dominant smokeless users and poly users compared with dominant cigarette users. Among the dominant cigarette users, the mean (SD) TD sum was 45.84 (16.66) and mean TD index was 0.12 (0.87). In comparison, dominant poly users had a mean (SD) TD sum of 39.59 (18.80) and TD index of 0.05 (0.81). They noted that these patterns remained stable across the survey waves despite users switching between TU profiles over time.

Overall, it was previously challenging to study TU profiles and TD over time due to the number of diverse tobacco products available, but the researchers’ updated study approach helped to estimate an individual’s use of various products; this helped to better capture the current TU profiles and their relationships with TD over time. With these TU profiles, they found that the TD levels varied, the highest being in dominant cigarette users. The researchers noted that their findings “may help clinicians and public health practitioners to identify targeted tobacco user groups for subsequent tobacco cessation counseling.

On the other hand, the researchers noted several limitations to their study, one being that they did not include the intensity and frequency of TU across groups, as they were measured differently across different tobacco products. Additionally, they may not be comparable due to vast heterogeneity in product characteristics, even for products in the same category.

The researchers concluded by urging future researchers to “examine patterns of TU profile transitions, including whether transitions differ by sociodemographic factors” and to analyze how the transitions impact TD levels in terms of other smoking behaviors, like quit attempts, relapses, and smoking cessation.

Reference
Li L, Yang C, Zhan S, et al. Longitudinal assessment of association between tobacco use and tobacco dependence among adults: latent class analysis of the population assessment of tobacco and health study waves 1-4. Nicotine Tob Res. Published online July 27, 2023. doi:10.1093/ntr/ntad114