Abstract
Background
As tobacco use patterns shift toward non-combustible tobacco products such as electronic cigarettes (ECs) and nicotine pouches, dual use with combustible products may undermine potential public health benefits. Herein we describe adult trends in exclusive and dual use of combustible and non-combustible products in the United States between 2015 and 2023.
Methods
We analyzed annual National Health and Interview Survey data including all adults with valid data on tobacco use, age, and sex. Joinpoint regression estimated annual percentage change (APC) and 95% CIs overall and by age group and sex. All tests were 2-tailed (P < .05).
Results
Exclusive combustible use remained 2-3 times higher than exclusive non-combustible use or dual use across all years. From 2015 and 2023, combustible use declined in both sexes, with the steepest drops in younger adults after 2020. Exclusive non-combustible use increased in all age groups after 2017, especially among younger adults and females. The most common dual use pattern was cigarettes plus ECs, which was stable over time.
Conclusions
Exclusive use of combustible tobacco, primarily cigarettes, is declining but remains the most common pattern, especially in older adults. In contrast, exclusive EC use is displacing combustible use among younger adults. Dual use patterns were low overall and stable across the study period.
Implications:
Exclusive use of combustible tobacco is declining while exclusive use of non-combustible products, particularly ECs, is increasing. The prevalence of combustible and non-combustible tobacco dual use was low and stable over time. Comparing the prevalence of any combustible tobacco use to exclusive non-combustible tobacco use suggests a population level shift toward lower-risk product use, indicating a potential net health benefit under accepted harm-reduction assumptions supported by biomarker evidence.
Introduction
Tobacco use remains a major contributor to morbidity and mortality in the United States. While both combustible and non-combustible tobacco products pose health risks to users, they confer vastly different risks.1-4 Factory made cigarettes are particularly harmful, responsible for about 480 000 deaths annually in the United States.4 Recent shifts in tobacco use patterns include declining rates of cigarette smoking and increased adoption of non-combustible products such as electronic cigarettes (ECs) and oral tobacco use.5-7 The differential risks of these products suggest that changing patterns of use could have important effects on future tobacco-related health outcomes.
While the prevalence of the more harmful combustible use is declining at the same time that the use of non-combustibles is increasing, reductions in tobacco-related morbidity and mortality may be limited if individuals use both combustible and non-combustible tobacco products.8-12 Biomarker studies suggest that smokers who switch completely to non-combustible tobacco products markedly reduce exposure to harmful and potentially harmful chemicals while those who dual use combustible and non-combustible tobacco do not get the same level of reduction in exposure to harmful chemicals.13-19 Several studies emphasize the importance of monitoring dual use, as it may undermine the risk reduction benefits of ECs and oral nicotine products.19-21
In this study, we describe population-level trends from 2015 to 2023 in exclusive and dual use of combustible and non-combustible tobacco among US adults, overall and stratified by age and sex. We also assess trends in exclusive and dual use of different product combinations, including cigarettes, other combustible tobacco products (ie, cigars, pipe and hookah tobacco combined), ECs and oral tobacco products. Finally, we introduce a harm-reduction-oriented surveillance indicator by comparing the prevalence of exclusive non-combustible tobacco use with the prevalence of any combustible tobacco use. Under commonly accepted harm-reduction assumptions supported by biomarker evidence,13-19 this prevalence-based ratio crudely describes population shifts toward lower-risk product use.
Methods
Study population and study design
In this cross-sectional trend-analysis study, we estimate the prevalence of current exclusive and dual use of combustible and non-combustible tobacco products using data from the National Health and Interview Survey (NHIS) from 2015 to 2023.22 The NHIS is a nationwide, annual, in-person survey of approximately 27 000 adults that collects data on a wide range of health topics including information on tobacco product usage of the civilian, non-institutionalized US population. The NHIS uses multistage probability sampling that incorporates stratification, clustering, and oversampling of some subpopulations (eg, Black, Hispanic, and Asian) in some years. Statistical weighting is used to generate representative national estimates.22 The US Census Bureau employs and trains field staff to conduct interviews. They follow standardized procedures and conduct the majority of interviews in person with some interviews done over the phone when an in person household interview is not possible.
In this pooled secondary data analysis, we chose to describe trends in exclusive and dual use of combustible and non-combustible tobacco product use beginning in 2015, when tobacco-use questions became consistent. In 2014, the NHIS survey combined the assessment of cigars and pipe use in a single question, whereas in 2015 separate questions were asked about cigars and pipe tobacco use. Previous studies have demonstrated that the formatting of questions about tobacco use can influence prevalence rates.23
Independent variables
The primary independent variable in this study was survey year, with 9 time points spanning from 2015 to 2023. In addition, we report tobacco use prevalence outcomes by age and sex. Age was measured in years and collapsed into 4 different age groups (ie, 18-24 years, 25-34 years, 35-54 years, and ≥55 years). Sex at birth was defined as either male or female. No other variables were included in this trend analysis.
Outcome variables
Table 1 provides the questions from the NHIS survey and definitions used for defining current prevalence of cigarettes, cigars, pipe/hookah tobacco, ECs, and smokeless tobacco. Current combustible tobacco use was defined as current use of any cigarettes, cigars, pipe and/or hookah tobacco. Current non-combustible use was defined as current use of any EC or smokeless tobacco. For each survey year, current tobacco use prevalence was categorized into 3 groups: (1) exclusive use of a combustible tobacco product; (2) exclusive use of a non-combustible tobacco product, and (3) dual use of both a combustible and non-combustible tobacco product. Exclusive use of a combustible tobacco was defined as current use of only combustible tobacco products, including cigarettes, cigars, pipe and/or hookah tobacco and no current use of non-combustible tobacco products. Exclusive use of a non-combustible tobacco was defined as current use of only non-combustible tobacco products, including EC or smokeless tobacco and no current use of combustible tobacco products. Dual use of combustible and non-combustible tobacco was defined as current reported use of both combustible tobacco and non-combustible products.
Table 1.
NHIS questions and definitions of current prevalence for cigarettes, cigars, pipe/hookah, e-cigarettes, and smokeless tobacco.
| Product | Text of questions | Current use definition a |
|---|---|---|
| Cigarettes |
|
Ever smoked 100 cigarettes in lifetime and use cigarettes every day or some days |
| Cigars |
|
Ever smoked a cigar even one time and now smokes a cigar every day or some days |
| Pipe/Hookah |
|
|
| E-cigarettes |
|
|
| Smokeless Tobacco |
|
|
Definition consistent throughout 2015-2023.
In addition, we also estimated the prevalence of exclusive use of individual combustible and non-combustible tobacco products as well as dual use as follows: (1) exclusive cigarette use; (2) exclusive non-cigarette combustible tobacco products defined as any current use of cigars, pipe tobacco, or hookah tobacco; (3) exclusive use of ECs; (4) exclusive use of oral tobacco defined as any use of chewing tobacco, moist snuff, or nicotine pouches; (5) dual use of cigarettes and ECs; (6) dual use of non-combustible tobacco products and ECs; (7) dual use of cigarettes and oral tobacco; and (8) dual use of non-combustible tobacco products and oral tobacco.
To assess the potential population health impact of shifting tobacco use patterns, we examined annual trends in the ratio of exclusive non-combustible tobacco prevalence to the prevalence of any combustible tobacco use (including exclusive and/or dual use), under the assumption that exclusive use of non-combustible tobacco use is less risky to health relative to combustible tobacco use based on biomarker evidence.13-19
Analyses
Descriptive statistics were generated using SAS 9.4 software (Cary, NC). For each NIHS survey year (2015-2023) we analyzed data using that year’s adult sampling weights and the NHIS-provided design variables for stratification and clustering (ie, strata and PSU). We estimated nationally representative annual prevalence and design-based standard errors using Taylor series linearization implemented using R survey package based on standard methodology (https://nhis.ipums.org/nhis/userNotes_variance.shtml). Because NHIS weights are cycle-specific, we did not pool the data across years for a single weighted analysis; instead, prevalence was estimated separately for each year using the corresponding annual weights. We then fit joinpoint regression models to the annual prevalence estimates and standard errors using weighted least squares estimates to estimate annual present change (APC) and identify potential inflection points. Joinpoint regression for trend analysis has been championed by the National Cancer Institute, which developed the Joinpoint regression software.
Joinpoint regression analyses were conducted to describe annual trends (2015-2023) of weighted prevalence estimates for the following categories: (1) exclusive use of any combustible tobacco product; (2) exclusive use of any non-combustible tobacco product; and (3) dual use of both combustible and non-combustible tobacco products. Additional joinpoint analyses were performed to assess prevalence trends of: (1) exclusive cigarette use; (2) exclusive use of non-cigarette combustible tobacco products (cigars, pipe tobacco, or hookah tobacco); (3) exclusive use of ECs; (4) exclusive oral tobacco use (chewing tobacco, moist snuff, or nicotine pouches); (5) dual use of cigarettes and ECs; (6) dual use of non-combustible tobacco products and ECs; (7) dual use of cigarettes and oral tobacco; and (8) dual use of non-combustible tobacco products and oral tobacco.
Joinpoint models do not incorporate individual-level covariates; therefore, our trend analysis did not “adjust” in the multivariable regression sense. Instead, to assess whether patterns differed across key demographic groups, we conducted joinpoint analyses stratified by age group and sex. For each analysis, we report the APC from 2015 to 2023 and corresponding 95% CIs, both overall (all age groups, sexes combined) and separately by age group (18-24, 25-34, 35-54, and ≥55) and sex (males, females). Joinpoint regression also distinguishes different time segments (ie, joinpoints) to identify significant differences between time points.24-26 We allowed for at most 1 change in trend (0 or 1 joinpoints). All statistical tests were 2-tailed and based on a P < .05 significance level. As in our previous work,7 statistical testing procedures followed Kim et al.25 as implemented in the NCI joinpoint regression software.24
Results
Table 2 shows the prevalence estimates for adults ≥18 years overall and stratified by age group and sex for exclusive combustible tobacco use. Table 3 presents corresponding estimates for exclusive non-combustible tobacco use, and Table 4 for dual use. Across all years, exclusive combustible tobacco use was 2-3 times more prevalent than exclusive non-combustible use or dual use. Exclusive non-combustible tobacco product use increased in all age groups after 2017, with the largest increases observed among younger adults and females.
Table 2.
Unweighted number (n) of people reporting exclusive combusted tobacco usea and weighted prevalence (%) estimates with 95% CIs, by age and sex, for participants in the US NHIS 2015-2023.
| Survey year |
2015 |
2016 |
2017 |
2018 |
2019 |
2020 |
2021 |
2022 |
2023 |
|---|---|---|---|---|---|---|---|---|---|
| Sex | Males and Females | ||||||||
| Age, >18 years | |||||||||
| Sample, n | 5245 | 5328 | 4123 | 3852 | 4375 | 3921 | 3618 | 3264 | 3312 |
| Prevalence, % | 15.7 | 15.7 | 14.6 | 14.3 | 14.2 | 13.2 | 12.3 | 12 | 11.2 |
| (95% CI) | (15.2 to 16.2) | (15.0 to 16.3) | (14.0 to 15.2) | (13.7 to 14.9) | (13.6 to 14.7) | (12.7 to 13.8) | (11.8 to 12.8) | (11.5 to 12.5) | (10.7 to 11.7) |
| Age, 18-24 years | |||||||||
| Sample, n | 425 | 414 | 264 | 179 | 180 | 115 | 105 | 61 | 57 |
| Prevalence, % | 14.2 | 13 | 10.6 | 8.4 | 7.5 | 6.5 | 5.2 | 3.8 | 2.9 |
| (95% CI) | (12.4 to 15.9) | (11.2 to 14.7) | (9.0 to 12.2) | (6.7 to 10.1) | (6.2 to 8.7) | (5.1 to 7.9) | (4.1 to 6.3) | (2.7 to 4.8) | (2.1 to 3.8) |
| Age, 25-34 years | |||||||||
| Sample, n | 1050 | 933 | 732 | 653 | 686 | 552 | 525 | 416 | 395 |
| Prevalence, % | 19.1 | 17.3 | 16.8 | 16.1 | 15.7 | 14.4 | 12.1 | 10.7 | 9.8 |
| (95% CI) | (17.6 to 20.5) | (15.8 to 18.8) | (15.4 to 18.2) | (14.8 to 17.5) | (14.3 to 17.0) | (13.0 to 15.9) | (10.9 to 13.3) | (9.6 to 11.8) | (8.7 to 10.8) |
| Age, 35-54 years | |||||||||
| Sample, n | 1976 | 1974 | 1476 | 1367 | 1612 | 1379 | 1316 | 1155 | 1137 |
| Prevalence, % | 17.9 | 18.4 | 16.6 | 16.7 | 17.1 | 15.7 | 15.1 | 15.1 | 13.9 |
| (95% CI) | (16.9 to 18.9) | (17.2 to 19.5) | (15.6 to 17.7) | (15.6 to 17.8) | (16.1 to 18.1) | (14.7 to 16.7) | (14.3 to 16.0) | (14.1 to 16.0) | (13.0 to 14.8) |
| Age, ≥55 years | |||||||||
| Sample, n | 1794 | 2007 | 1651 | 1653 | 1897 | 1875 | 1672 | 1632 | 1723 |
| Prevalence, % | 12.5 | 13.3 | 13 | 13.2 | 13 | 12.6 | 12.1 | 12.4 | 12.2 |
| (95% CI) |
(11.7 to 13.3) |
(12.5 to 14.1) |
(12.2 to 13.8) |
(12.4 to 13.9) |
(12.3 to 13.7) |
(11.9 to 13.4) |
(11.4 to 12.7) |
(11.7 to 13.1) |
(11.5 to 12.9) |
|
Survey Year |
2015 |
2016 |
2017 |
2018 |
2019 |
2020 |
2021 |
2022 |
2023 |
|---|---|---|---|---|---|---|---|---|---|
| Sex | Males only | ||||||||
| Age, >18 years | |||||||||
| Sample, n | 2788 | 2859 | 2274 | 2099 | 2400 | 2207 | 2043 | 1851 | 1909 |
| Prevalence, % | 18.1 | 19 | 17.8 | 17.2 | 16.5 | 15.8 | 15.2 | 14.5 | 14.1 |
| (95% CI) | (17.3 to 19.0) | (18.0 to 20.0) | (16.9 to 18.6) | (16.3 to 18.1) | (15.7 to 17.3) | (15.0 to 16.7) | (14.5 to 16.0) | (13.8 to 15.2) | (13.3 to 14.8) |
| Age, 18-24 years | |||||||||
| Sample, n | 220 | 215 | 154 | 106 | 91 | 65 | 66 | 38 | 40 |
| Prevalence, % | 15.4 | 14 | 12.6 | 9.8 | 7.1 | 7 | 6.6 | 4.6 | 4.3 |
| (95% CI) | (11.5 to 16.6) | (11.5 to 16.6) | (10.1 to 15.1) | (7.5 to 12.1) | (5.4 to 8.8) | (5.0 to 9.1) | (4.8 to 8.4) | (3.0 to 6.2) | (2.7 to 5.8) |
| Age, 25-34 years | |||||||||
| Sample, n | 553 | 478 | 407 | 336 | 370 | 331 | 312 | 251 | 236 |
| Prevalence, % | 23 | 21.3 | 20.6 | 18.2 | 18.4 | 17.9 | 15.7 | 13.1 | 12.8 |
| (95% CI) | (20.5 to 25.4) | (19.0 to 23.7) | (18.4 to 22.7) | (16.2 to 20.3) | (16.3 to 20.6) | (15.6 to 20.1) | (13.8 to 17.6) | (11.3 to 14.8) | (11.1 to 14.5) |
| Age, 35-54 years | |||||||||
| Sample, n | 1021 | 1046 | 819 | 737 | 861 | 777 | 759 | 675 | 653 |
| Prevalence, % | 19 | 21.4 | 20.2 | 19.8 | 19.3 | 18.5 | 18.6 | 18.3 | 16.9 |
| (95% CI) | (17.6 to 20.4) | (19.7 to 23.0) | (18.7 to 21.7) | (18.1 to 21.4) | (17.9 to 20.8) | (17.0 to 20.1) | (17.2 to 19.9) | (16.8 to 19.8) | (15.5 to 18.3) |
| Age, ≥55 years | |||||||||
| Sample, n | 994 | 1120 | 894 | 920 | 1078 | 1034 | 906 | 887 | 980 |
| Prevalence, % | 15.8 | 17.2 | 15.9 | 16.8 | 16.2 | 15.3 | 14.8 | 15 | 15.4 |
| (95% CI) | (14.5 to 17.1) | (16.0 to 18.4) | (14.6 to 17.1) | (15.6 to 18.0) | (15.1 to 17.3) | (14.2 to 16.4) | (13.8 to 15.9) | (13.9 to 16.0) | (14.3 to 16.5) |
| Survey Year |
2015 |
2016 |
2017 |
2018 |
2019 |
2020 |
2021 |
2022 |
2023 |
|---|---|---|---|---|---|---|---|---|---|
| Sex | Females only | ||||||||
| Age, >18 years | |||||||||
| Sample, n | 2457 | 2469 | 1849 | 1753 | 1975 | 1714 | 1575 | 1413 | 1403 |
| Prevalence, % | 13.4 | 12.6 | 11.6 | 11.60% | 12 | 10.8 | 9.5 | 9.6 | 8.5 |
| (95% CI) | (12.8 to 14.1) | (11.9 to 13.3) | (10.9 to 12.4) | (10.9 to 12.3) | (11.3 to 12.6) | (10.2 to 11.5) | (8.9 to 10.0) | (8.9 to 10.2) | (7.9 to 9.0) |
| Age, 18-24 years | |||||||||
| Sample, n | 205 | 199 | 110 | 73 | 89 | 50 | 39 | 23 | 17 |
| Prevalence, % | 12.9 | 11.9 | 8.6 | 7 | 7.8 | 6 | 3.8 | 2.9 | 1.6 |
| (95% CI) | (10.4 to 15.4) | (9.7 to 14.0) | (6.7 to 10.6) | (4.8 to 9.2) | (5.9 to 9.7) | (4.1 to 7.8) | (2.5 to 5.1) | (1.5 to 4.3) | (0.8 to 2.4) |
| Age, 25-34 years | |||||||||
| Sample, n | 497 | 455 | 325 | 317 | 316 | 221 | 213 | 165 | 159 |
| Prevalence, % | 15.3 | 13.4 | 13 | 14.1 | 12.9 | 11 | 8.5 | 8.3 | 6.7 |
| (95% CI) | (13.6 to 17.0) | (11.7 to 15.0) | (11.4 to 14.7) | (12.2 to 16.1) | (11.4 to 14.5) | (9.4 to 12.7) | (7.3 to 9.8) | (6.8 to 9.7) | (5.6 to 7.9) |
| Age, 35-54 years | |||||||||
| Sample, n | 955 | 928 | 657 | 630 | 751 | 602 | 557 | 480 | 484 |
| Prevalence, % | 16.8 | 15.5 | 13.2 | 13.7 | 14.9 | 13 | 11.8 | 11.9 | 10.8 |
| (95% CI) | (15.4 to 18.1) | (14.1 to 16.8) | (11.9 to 14.5) | (12.4 to 15.0) | (13.7 to 16.2) | (11.7 to 14.2) | (10.7 to 12.9) | (10.8 to 13.1) | (9.7 to 11.9) |
| Age, ≥55 years | |||||||||
| Sample, n | 800 | 887 | 757 | 733 | 819 | 841 | 766 | 745 | 743 |
| Prevalence, % | 9.7 | 10 | 10.50 | 10 | 10.3 | 10.4 | 9.7 | 10.1 | 9.3 |
| (95% CI) | (8.9 to 10.6) | (9.1 to 10.8) | (9.6 to 11.5) | (9.1 to 10.9) | (9.5 to 11.2) | (9.5 to 11.2) | (8.9 to 10.4) | (9.3 to 10.9) | (8.5 to 10.1) |
Exclusive combustible tobacco use includes using any of the following tobacco products: cigarettes, cigars, pipe tobacco, hookah tobacco.
Table 3.
Unweighted number (n) of people reporting exclusive non-combustible tobacco usea and weighted prevalence (%) estimates with 95% CIs, by age and sex, for participants in the US NHIS 2015-2023.
| Survey year |
2015 |
2016 |
2017 |
2018 |
2019 |
2020 |
2021 |
2022 |
2023 |
|---|---|---|---|---|---|---|---|---|---|
| Sex | Males and Females | ||||||||
| Age, >18 years | |||||||||
| Sample, n | 753 | 923 | 691 | 760 | 1131 | 974 | 1047 | 1127 | 1343 |
| Prevalence, % | 2.7 | 2.8 | 2.6 | 3.2 | 4.1 | 3.9 | 4.2 | 5.2 | 5.6 |
| (95% CI) | (2.4 to 3.0) | (2.5 to 3.0) | (2.3 to 2.8) | (2.9 to 3.5) | (3.8 to 4.3) | (3.6 to 4.2) | (3.9 to 4.5) | (4.8 to 5.6) | (5.2 to 5.9) |
| Age, 18-24 years | |||||||||
| Sample, n | 98 | 110 | 101 | 120 | 153 | 120 | 167 | 194 | 198 |
| Prevalence, % | 3.9 | 3.6 | 4.3 | 5.9 | 6.9 | 6.6 | 8.6 | 11.9 | 10.9 |
| (95% CI) | (3.0 to 4.9) | (2.8 to 4.4) | (3.1 to 5.4) | (4.6 to 7.2) | (5.7 to 8.2) | (5.2 to 8.0) | (7.1 to 10.0) | (10.0 to 13.7) | (9.2 to 12.5) |
| Age, 25-34 years | |||||||||
| Sample, n | 170 | 167 | 146 | 164 | 256 | 242 | 253 | 304 | 386 |
| Prevalence, % | 3.1 | 3 | 3.5 | 4.1 | 5.6 | 5.9 | 5.8 | 8.5 | 9.4 |
| (95% CI) | (2.5 to 3.7) | (2.4 to 3.6) | (2.8 to 4.1) | (3.4 to 4.8) | (4.7 to 6.4) | (5.0 to 6.8) | (5.0 to 6.6) | (7.4 to 9.5) | (8.3 to 10.4) |
| Age, 35-54 years | |||||||||
| Sample, n | 297 | 387 | 256 | 260 | 439 | 345 | 372 | 375 | 454 |
| Prevalence, % | 3.1 | 3.7 | 2.7 | 3.4 | 4.6 | 4.1 | 4.4 | 5 | 5.5 |
| (95% CI) | (2.6 to 3.5) | (3.2 to 4.1) | (2.3 to 3.1) | (2.9 to 3.8) | (4.1 to 5.0) | (3.6 to 4.6) | (3.9 to 5.0) | (4.5 to 5.6) | (4.9 to 6.1) |
| Age, ≥55 years | |||||||||
| Sample, n | 188 | 259 | 188 | 216 | 293 | 267 | 255 | 254 | 305 |
| Prevalence, % | 1.7 | 1.5 | 1.5 | 1.8 | 2 | 2 | 2 | 1.9 | 2.2 |
| (95% CI) |
(1.3 to 2.0) |
(1.3 to 1.7) |
(1.2 to 1.7) |
(1.5 to 2.1) |
(1.7 to 2.3) |
(1.6 to 1.3) |
(1.7 to 2.3) |
(1.7 to 2.2) |
(1.9 to 2.5) |
|
Survey Year |
2015 |
2016 |
2017 |
2018 |
2019 |
2020 |
2021 |
2022 |
2023 |
|---|---|---|---|---|---|---|---|---|---|
| Sex | Males only | ||||||||
| Age, >18 years | |||||||||
| Sample, n | 587 | 726 | 517 | 584 | 822 | 698 | 710 | 691 | 852 |
| Prevalence, % | 4.4 | 4.5 | 3.9 | 5.2 | 6.1 | 5.7 | 6 | 6.5 | 7.3 |
| (95% CI) | (3.9 to 5.0) | (4.0 to 4.9) | (3.5 to 4.4) | (4.7 to 5.8) | (5.6 to 6.6) | (5.2 to 6.2) | (5.4 to 6.5) | (6.0 to 7.1) | (6.7 to 7.9) |
| Age, 18-24 years | |||||||||
| Sample, n | 81 | 92 | 74 | 88 | 98 | 70 | 91 | 99 | 108 |
| Prevalence, % | 6.9 | 6.2 | 6.1 | 8.6 | 8.8 | 7.7 | 9.3 | 12 | 11.6 |
| (95% CI) | (5.1 to 8.8) | (4.6 to 7.7) | (4.4 to 7.7) | (6.4 to 10.8) | (6.8 to 10.8) | (5.6 to 9.8) | (7.2 to 11.4) | (9.4 to 14.6) | (9.4 to 13.9) |
| Age, 25-34 years | |||||||||
| Sample, n | 142 | 139 | 102 | 126 | 191 | 174 | 169 | 179 | 238 |
| Prevalence, % | 5.5 | 4.9 | 4.6 | 6.6 | 8.3 | 8.6 | 8 | 10.1 | 12.1 |
| (95% CI) | (4.4 to 6.7) | (3.8 to 6.0) | (3.6 to 5.7) | (5.3 to 7.9) | (6.9 to 9.8) | (7.1 to 10.0) | (6.7 to 9.3) | (8.5 to 11.7) | (10.3 to 13.8) |
| Age, 35-54 years | |||||||||
| Sample, n | 242 | 297 | 203 | 207 | 329 | 259 | 261 | 246 | 286 |
| Prevalence, % | 4.8 | 5.6 | 4.5 | 5.6 | 7.1 | 6.4 | 6.7 | 6.9 | 7.4 |
| (95% CI) | (4.0 to 5.7) | (4.7 to 6.4) | (3.7 to 5.2) | (4.7 to 6.5) | (6.2 to 7.9) | (5.5 to 7.3) | (5.8 to 7.7) | (5.9 to 7.8) | (6.4 to 8.4) |
| Age, ≥55 years | |||||||||
| Sample, n | 122 | 198 | 138 | 163 | 204 | 195 | 189 | 167 | 220 |
| Prevalence, % | 2.5 | 2.5 | 2.3 | 3 | 3.2 | 3 | 3.2 | 2.8 | 3.4 |
| (95% CI) | (1.8 to 3.2) | (2.1 to 2.9) | (1.8 to 2.9) | (2.4 to 3.6) | (2.6 to 3.7) | (2.5 to 3.6) | (2.6 to 3.7) | (2.3 to 3.3) | (2.9 to 4.0) |
| Survey Year |
2015 |
2016 |
2017 |
2018 |
2019 |
2020 |
2021 |
2022 |
2023 |
|---|---|---|---|---|---|---|---|---|---|
| Sex | Females only | ||||||||
| Age, >18 years | |||||||||
| Sample, n | 166 | 197 | 174 | 176 | 309 | 276 | 337 | 436 | 491 |
| Prevalence, % | 1 | 1.2 | 1.3 | 1.3 | 2.1 | 2.2 | 2.6 | 3.9 | 3.9 |
| (95% CI) | (0.8 to 1.2) | (1.0 to 1.4) | (1.0 to 1.6) | (1.1 to 1.5) | (1.8 to 2.4) | (1.8 to 2.5) | (2.2 to 2.9) | (3.5 to 4.4) | (3.5 to 4.3) |
| Age, 18-24 years | |||||||||
| Sample, n | 17 | 18 | 27 | 32 | 55 | 50 | 76 | 95 | 90 |
| Prevalence, % | 0.9 | 1 | 2.5 | 3.1 | 5.1 | 5.6 | 7.80% | 11.7 | 10.1 |
| (95% CI) | (0.4 to 1.4) | (0.5 to 1.5) | (0.8 to 4.1) | (1.9 to 4.4) | (3.6 to 6.6) | (3.7 to 7.5) | (5.9 to 9.8) | (9.0 to 14.5) | (7.8 to 12.4) |
| Age, 25-34 years | |||||||||
| Sample, n | 28 | 28 | 44 | 38 | 65 | 68 | 84 | 125 | 148 |
| Prevalence, % | 0.8 | 1.1 | 2.3 | 1.6 | 2.8 | 3.2 | 3.6 | 6.9 | 6.7 |
| (95% CI) | (0.4 to 1.1) | (0.6 to 1.6) | (1.5 to 3.2) | (1.0 to 2.2) | (1.9 to 3.6) | (2.3 to 4.1) | (2.8 to 4.5) | (5.5 to 8.2) | (5.5 to 7.9) |
| Age, 35-54 years | |||||||||
| Sample, n | 55 | 90 | 53 | 53 | 110 | 86 | 111 | 129 | 168 |
| Prevalence, % | 1.3 | 1.9 | 1 | 1.2 | 2.2 | 1.9 | 2.2 | 3.3 | 3.7 |
| (95% CI) | (0.9 to 1.8) | (1.3 to 2.4) | (0.7 to 1.3) | (0.8 to 1.6) | (1.7 to 2.6) | (1.4 to 2.3) | (1.7 to 2.7) | (2.6 to 3.9) | (3.0 to 4.3) |
| Age, ≥55 years | |||||||||
| Sample, n | 66 | 61 | 50 | 53 | 79 | 72 | 66 | 87 | 85 |
| Prevalence, % | 0.9 | 0.6 | 0.7 | 0.7 | 1 | 1 | 1 | 1.1 | 1.1 |
| (95% CI) | (0.6 to 1.2) | (0.4 to 0.8) | (0.5 to 0.9) | (0.5 to 0.9) | (0.7 to 1.2) | (0.7 to 1.4) | (0.7 to 1.2) | (0.9 to 1.4) | (0.8 to 1.4) |
Exclusive non-combustible tobacco use includes using any of the following tobacco products: e-cigarettes, chewing tobacco, moist snuff, nicotine pouches.
Table 4.
Unweighted number (n) of people reporting dual use of combusted and non-combusted tobaccoa and weighted prevalence (%) estimates with 95% CIs, by age and sex, for participants in the US NHIS 2015-2023.
| Survey year |
2015 |
2016 |
2017 |
2018 |
2019 |
2020 |
2021 |
2022 |
2023 |
|---|---|---|---|---|---|---|---|---|---|
| Sex | Males and Females | ||||||||
| Age, >18 years | |||||||||
| Sample, n | 905 | 862 | 596 | 603 | 712 | 462 | 585 | 607 | 674 |
| Prevalence, % | 2.8 | 2.7 | 2.1 | 2.3 | 2.6 | 2 | 2.3 | 2.6 | 2.8 |
| (95% CI) | (2.5 to 3.1) | (2.4 to 2.9) | (1.9 to 2.3) | (2.0 to 2.5) | (2.3 to 2.8) | (1.7 to 2.2) | (2.0 to 2.5) | (2.4 to 2.9) | (2.5 to 3.0) |
| Age, 18-24 years | |||||||||
| Sample, n | 122 | 122 | 90 | 71 | 80 | 71 | 65 | 68 | 62 |
| Prevalence, % | 4.1 | 4 | 3.4 | 2.8 | 3.8 | 4.5 | 3.3 | 4.2 | 3 |
| (95% CI) | (3.1 to 5.1) | (2.8 to 5.1) | (2.6 to 4.3) | (2.0 to 3.7) | (2.7 to 4.8) | (3.2 to 5.7) | (2.4 to 4.2) | (3.1 to 5.3) | (2.2 to 3.8) |
| Age, 25-34 years | |||||||||
| Sample, n | 200 | 201 | 169 | 170 | 214 | 128 | 172 | 168 | 204 |
| Prevalence, % | 3.4 | 3.9 | 3.7 | 4 | 4.6 | 3 | 4.1 | 4.5 | 5.4 |
| (95% CI) | (2.7 to 4.1) | (3.2 to 4.7) | (3.1 to 4.4) | (3.3 to 4.7) | (3.9 to 5.3) | (2.3 to 3.6) | (3.4 to 4.8) | (3.7 to 5.3) | (4.6 to 6.3) |
| Age, 35-54 years | |||||||||
| Sample, n | 373 | 329 | 204 | 236 | 281 | 187 | 218 | 241 | 264 |
| Prevalence, % | 3.4 | 2.9 | 2.1 | 2.6 | 2.9 | 2.1 | 2.4 | 3 | 3.4 |
| (95% CI) | (2.9 to 3.8) | (2.5 to 3.3) | (1.7 to 2.4) | (2.2 to 3.0) | (2.5 to 3.3) | (1.8 to 2.5) | (2.1 to 2.8) | (2.5 to 2.4) | (2.9 to 3.9) |
| Age, ≥55 years | |||||||||
| Sample, n | 210 | 210 | 133 | 126 | 137 | 76 | 130 | 130 | 144 |
| Prevalence, % | 1.5 | 1.3 | 1 | 0.9 | 0.9 | 0.6 | 1 | 1 | 1 |
| (95% CI) |
(1.2 to 1.8) |
(1.1 to 1.6) |
(0.8 to 1.2) |
(0.7 to 1.1) |
(0.7 to 1.1) |
(0.4 to 0.7) |
(0.8 to 1.2) |
(0.8 to 1.3) |
(0.8 to 1.2) |
|
Survey Year |
2015 |
2016 |
2017 |
2018 |
2019 |
2020 |
2021 |
2022 |
2023 |
|---|---|---|---|---|---|---|---|---|---|
| Sex | Males only | ||||||||
| Age, >18 years | |||||||||
| Sample, n | 558 | 555 | 396 | 435 | 471 | 329 | 357 | 412 | 436 |
| Prevalence, % | 3.9 | 3.7 | 3.1 | 3.4 | 3.6 | 3 | 3 | 3.9 | 3.9 |
| (95% CI) | (3.4 to 4.4) | (3.2 to 4.2) | (2.7 to 3.4) | (3.0 to 3.8) | (3.2 to 4.0) | (2.6 to 3.4) | (2.6 to 3.3) | (3.4 to 4.3) | (3.4 to 4.3) |
| Age, 18-24 years | |||||||||
| Sample, n | 88 | 94 | 66 | 50 | 50 | 56 | 37 | 51 | 42 |
| Prevalence, % | 6.2 | 6.4 | 5 | 3.6 | 4.7 | 7.4 | 3.9 | 6.3 | 3.9 |
| (95% CI) | (4.5 to 7.9) | (4.2 to 8.6) | (3.6 to 6.5) | (2.3 to 5.0) | (3.1 to 6.4) | (5.2 to 9.6) | (2.5 to 5.4) | (4.4 to 8.1) | (2.7 to 5.2) |
| Age, 25-34 years | |||||||||
| Sample, n | 145 | 142 | 128 | 131 | 152 | 98 | 114 | 125 | 144 |
| Prevalence, % | 5.1 | 5.8 | 5.8 | 6.4 | 6.4 | 4.6 | 5.9 | 7.1 | 7.8 |
| (95% CI) | (3.9 to 6.3) | (4.5 to 7.1) | (4.6 to 7.0) | (5.2 to 7.7) | (5.2 to 7.5) | (3.5 to 5.6) | (4.7 to 7.1) | (5.7 to 8.6) | (6.4 to 9.2) |
| Age, 35-54 years | |||||||||
| Sample, n | 215 | 212 | 132 | 174 | 190 | 132 | 134 | 155 | 167 |
| Prevalence, % | 4.5 | 3.9 | 2.8 | 3.9 | 4.1 | 3.1 | 3.1 | 4 | 4.7 |
| (95% CI) | (3.7 to 5.2) | (3.2 to 4.5) | (2.3 to 3.4) | (3.2 to 4.7) | (3.4 to 4.7) | (2.5 to 3.8) | (2.5 to 3.6) | (3.3 to 4.6) | (3.8 to 5.5) |
| Age, ≥55 years | |||||||||
| Sample, n | 110 | 107 | 70 | 80 | 79 | 43 | 72 | 81 | 83 |
| Prevalence, % | 1.9 | 1.5 | 1.2 | 1.3 | 1.3 | 0.8 | 1.1 | 1.4 | 1.3 |
| (95% CI) | (1.4 to 2.4) | (1.2 to 1.8) | (0.9 to 1.5) | (1.0 to 1.6) | (0.9 to 1.6) | (0.5 to 1.0) | (0.8 to 1.4) | (1.0 to 1.8) | (1.0 to 1.6) |
| Survey Year |
2015 |
2016 |
2017 |
2018 |
2019 |
2020 |
2021 |
2022 |
2023 |
|---|---|---|---|---|---|---|---|---|---|
| Sex | Females only | ||||||||
| Age, >18 years | |||||||||
| Sample, n | 347 | 307 | 200 | 168 | 241 | 133 | 228 | 195 | 238 |
| Prevalence, % | 1.8 | 1.7 | 1.3 | 1.2 | 1.6 | 0.9 | 1.6 | 1.4 | 1.7 |
| (95% CI) | (1.5 to 2.1) | (1.4 to 1.9) | (1.0 to 1.5) | (1.0 to 1.4) | (1.4 to 1.9) | (0.7 to 1.1) | (1.3 to 1.8) | (1.2 to 1.7) | (1.4 to 1.9) |
| Age, 18-24 years | |||||||||
| Sample, n | 34 | 28 | 24 | 21 | 30 | 15 | 28 | 17 | 20 |
| Prevalence, % | 2 | 1.5 | 1.8 | 2 | 2.7 | 1.5 | 2.6 | 2 | 2 |
| (95% CI) | (1.0 to 2.9) | (0.8 to 2.3) | (1.0 to 2.6) | (0.9 to 3.0) | (1.6 to 3.9) | (0.6 to 2.3) | (1.5 to 3.7) | (0.9 to 3.1) | (1.1 to 3.0) |
| Age, 25-34 years | |||||||||
| Sample, n | 55 | 59 | 41 | 39 | 62 | 30 | 58 | 43 | 60 |
| Prevalence, % | 1.8 | 2.1 | 1.7 | 1.7 | 2.9 | 1.4 | 2.2 | 1.9 | 3 |
| (95% CI) | (1.1 to 2.5) | (1.2 to 2.9) | (1.1 to 2.3) | (1.1 to 2.3) | (2.1 to 3.6) | (0.8 to 2.0) | (1.5 to 2.9) | (1.3 to 2.5) | (2.2 to 3.9) |
| Age, 35-54 years | |||||||||
| Sample, n | 158 | 117 | 72 | 62 | 91 | 55 | 84 | 86 | 97 |
| Prevalence, % | 2.3 | 2 | 1.3 | 1.3 | 1.8 | 1.2 | 1.8 | 2 | 2.1 |
| (95% CI) | (1.8 to 2.9) | (1.6 to 2.5) | (0.9 to 1.7) | (0.9 to 1.7) | (1.3 to 2.2) | (0.8 to 1.6) | (1.3 to 2.8) | (1.5 to 2.5) | (1.6 to 2.6) |
| Age, ≥55 years | |||||||||
| Sample, n | 100 | 103 | 63 | 46 | 58 | 33 | 58 | 49 | 61 |
| Prevalence, % | 1.2 | 1.2 | 0.8 | 0.6 | 0.6 | 0.4 | 0.9 | 0.7 | 0.7 |
| (95% CI) | (0.9 to 1.6) | (0.9 to 1.5) | (0.6 to 1.1) | (0.4 to 0.8) | (0.4 to 0.8) | (0.2 to 0.6) | (0.6 to 1.1) | (0.5 to 0.9) | (0.5 to 0.9) |
Dual use of combusted and non-combusted tobacco means use of any of the following combusted tobacco products: cigarettes, cigars, pipe tobacco and/or hookah tobacco and any of the following non-combusted tobacco products: e-cigarettes, chewing tobacco, moist snuff, and/or nicotine pouches.
Tables S1-S8 present the prevalence estimates for exclusive and dual use of specific tobacco products and product combinations (eg, cigarettes, other combustible tobacco products, ECs, and oral tobacco products). The most prominent changes over the study period were declining exclusive cigarette prevalence (Table S1) and increasing exclusive EC use (Table S3). The combination of cigarettes and EC was the most common dual-use pattern observed, although unchanged between 2015 through 2023. Figure 1 graphically displays the weighted overall prevalence trends for exclusive cigarette use, exclusive EC use, and dual use during the study period. Tables S4 and S6 shows exclusive and dual use of cigarettes and oral tobacco. Other dual use combinations had low prevalence rates and remained stable from 2015 to 2023.
Figure 1.
Overall weighted prevalence estimates of exclusive cigarette use, exclusive e-cigarette use, or dual use of both products among participants of the NHIS survey 2015 through 2023.
Figure 2 displays the joinpoint regression plots overall and by age and sex for exclusive combustible tobacco product use. As shown in Figure 2, overall exclusive combustible tobacco prevalence declined 3.6% annually (95% CI = −4.6% to −2.6%), but the magnitude of change was more pronounced in younger age groups, especially after 2020. Figure 3 shows the joinpoint regression plots overall and by age and sex for exclusive non-combustible tobacco use. As shown in Figure 3, exclusive non-combustible tobacco prevalence increased at an APC of 10.5% (95% CI = 7.2% to 13.9%), with statistically significant increases observed in all age groups and in males and females. Figure 4 shows the joinpoint regression plots overall and by age and sex for dual use. As shown in Figure 4, the APC for dual use was stable overtime.
Figure 2.
Joinpoint regression plots for the prevalencea of exclusive combustible tobacco useb by age and sex for participants in the US NHIS from 2015 through 2023. aPrevalence estimates based on weighted data. bExclusive combustible tobacco use means only use of any of the following tobacco products: cigarettes, cigars, pipe tobacco, or hookah tobacco. Bold indicates significant at the P < .05 level.
Figure 3.
Joinpoint regression plots for the prevalencea of exclusive non-combustible tobacco useb by age and sex for participants in the US NHIS from 2015 through 2023. aPrevalence estimates based on weighted data. bExclusive use of non-combustible tobacco includes only use of any of the following tobacco products: ECs, chewing tobacco, moist snuff, or nicotine pouches. Bold indicates significant at the P < .05 level.
Figure 4.
Joinpoint regression plots for the prevalencea of dual use of combusted and non-combusted tobacco productsb by age and sex for participants in the US NHIS from 2015 through 2023. aPrevalence estimates based on weighted data. bDual use includes reported use of any combusted tobacco product (ie, cigarettes, cigars, pipe tobacco, hookah tobacco) AND any use of a non-combusted tobacco product (ie, ECs, chewing tobacco, moist snuff, or nicotine pouches). Bold indicates significant at the P < .05 level.
Figures S1-S8 display the joinpoint regression plots describing trends in exclusive and dual use of different combustible and non-combustible tobacco products. The most prominent changes were decreasing cigarette use and increasing EC use. Between 2019 and 2023, exclusive cigarette prevalence declined 6.6% (95% CI = −14.2% to −3.6%) overall, with more pronounced reductions in younger age groups (ie, 18-24 years) compared to older age groups (≥35 years) (Figure S1). Between 2015 and 2023, exclusive EC use increased at an APC of 18.8% (95 CI = 13.0% to 24.9%) with statistically significant increases observed in all age groups and in males and females (Figure S3).
Figure 5 shows the joinpoint regression plots revealing trends in the ratio of exclusive non-combustible tobacco use prevalence to the prevalence of any combustible overall and by age group, sex, and survey year. Overall, the prevalence of exclusive non-combustible tobacco use compared to any combustible tobacco use increased over time in all age and sex groups, reflecting a significant shift to exclusive non-combustible tobacco products compared to any combustible tobacco product use. As shown in Figure 5, the APC percentages were most pronounced among younger age groups (ie, 18-34 years) and appeared to have increased more among females after 2018, reflective of the growing popularity of exclusive EC use among females. Figure S9 presents trends in the ratio of exclusive EC prevalence to the prevalence of exclusive cigarette use and dual use of cigarettes and ECs. The pattern of results is like those shown in Figure 5.
Figure 5.
Joinpoint regression plots show trends in the ratio of exclusive use of non-combustible tobacco usea to any use of combusted tobacco productsb by age and sex for participants in the US NHIS from 2015 through 2023. aExclusive use of non-combustible tobacco includes only use of any of the following tobacco products: ECs, chewing tobacco, moist snuff, or nicotine pouches. bAny use of a combustible tobacco product includes exclusive or dual use that involves any of the following tobacco products: cigarettes, cigars, pipe tobacco, or hookah tobacco. Bold indicates significant at the P < .05 level.
Discussion
This study describes recent population-level trends in exclusive and dual use of combustible and non-combustible tobacco products among US adults, with analyses stratified by age and sex, as well as by different tobacco product combinations. We found that exclusive use of combustible tobacco, primarily cigarettes, declined between 2015 and 2023 but remained the predominant use pattern overall, particularly among older adults. In contrast, exclusive use of non-combustible tobacco products rose across all age groups after 2017, and by 2023, exclusive EC use was more common than combustible tobacco among adults aged 18-24. Dual use of combustible and non-combustible products remained relatively stable between 2015 and 2023 but was more prevalent among males and younger adults (18-34 years). These findings underscore the importance of monitoring evolving tobacco use patterns, as they have potential implications for population-level health risks, harm reduction strategies, and future policy effects.
The findings from this descriptive study indicate that US adult tobacco use is increasingly shifting away from combustible tobacco products, mostly cigarettes, toward exclusive use of non-combustible tobacco products, primarily ECs.7 While non-combustible tobacco product use can cause nicotine dependence and is not without health risks, the absence of burned tobacco reduces exposure to harmful and potentially harmful chemicals.13-19 Consistent with these harm-reduction assumptions, the increasing prevalence of exclusive non-combustible tobacco product use compared to any combustible tobacco product use as displayed in Figure 5 serves as a crude descriptive measure of population-level progress toward lower-risk product use. While the decline in combustible tobacco product use, either exclusively or as part of dual use, is encouraging news, the rate of decline was much less among those ≥35 years of age, highlighting the need for enhanced efforts to discourage the use of smoked tobacco by older Americans.27
While this study provides a descriptive assessment of population-level trends in tobacco product use among US adults, readers should not overinterpret these findings for several reasons. First, we utilized NHIS-defined prevalence of current tobacco product use, which differed for established current cigarette smoking (ie, 100 cigarettes over lifetime and currently smoking every day or some days) compared to non-cigarette tobacco products, which assessed prevalence based on ever use and current use every day and some days. The different definitions may affect comparability across products and classification of exclusive and dual use. In addition, prevalence measures do not capture information on the frequency and duration of tobacco product use, both of which are important determinants of future health risks.4 Second, self-reported tobacco use is prone to information bias and misclassification. For example, the negative stigma associated with smoking cigarettes has been found to be associated with underreporting smoking in surveys.28 Also, not all respondents will be aware of the different tobacco products asked about in the survey, which may contribute to measurement error. Third, in this study we only report on cross-sectional population level prevalence rates and do not consider individual patterns of tobacco product use, which are likely complex with people transitioning over time between different tobacco product use states, including using more than one product at a time (ie, dual use).10-12,29 For example, studies of people who report currently using cigarettes have found that most are still smoking cigarettes in the next year or even 5 years later.30,31 This finding might not apply to those using non-cigarette tobacco products, suggesting that different tobacco products might convey differential risks of abuse liability. Fourth, this study did not consider dual use of cannabis smoking, which could impact future health risks. There is an association between tobacco use and cannabis use and recent studies have found exclusive use of nicotine vaping is associated with the smoking of cannabis, especially in young adults.32 Fifth, there are inherent limitations of the joinpoint analyses presented in this paper, since we report on only 9 data points. Results may change as additional data points become available. For example, between 2015 and 2023 the prevalence of nicotine pouch use was low. Recent studies have found that the use of nicotine pouches has doubled in the past 2 years so it will be important to see how growth in this product category influences trends in tobacco use patterns.33 Sixth, the findings from this study may not be generalizable to countries outside the United States.
In summary, the findings from this study indicate a positive trend in tobacco use in the United States between 2015 and 2023, characterized by reduced use of smoked tobacco products, a pattern most evident among younger adults. Continuous surveillance of trends in the ratio of the prevalence of exclusive non-combustible tobacco use to the prevalence of any combustible tobacco use provides a straightforward, albeit crude, metric for assessing population-level progress in reducing the future harms associated with tobacco product use.
Supplementary Material
Acknowledgments
This research was supported by grant U54CA229974 from the United States National Cancer Institute (NCI) and United States Food and Drug Administration’s Center for Tobacco (FDA); and P01 CA200512 from the NCI. The funders had no role in the design or conduct of the study; collection, management, analysis, or interpretation of the data; preparation, review, or approval of the manuscript; or decision to submit the manuscript for publication. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH or the FDA.
Contributor Information
Kenneth Michael Cummings, Department of Psychiatry and Behavioral Sciences, Medical University of South Carolina, Charleston, SC, United States.
Avery Roberson, Department of Psychiatry and Behavioral Sciences, Medical University of South Carolina, Charleston, SC, United States.
Colin W Randol, Department of Psychiatry and Behavioral Sciences, Medical University of South Carolina, Charleston, SC, United States.
Tracy T Smith, Department of Psychiatry and Behavioral Sciences, Medical University of South Carolina, Charleston, SC, United States.
David T Levy, Department of Oncology, Lombardi Comprehensive Cancer Center, Georgetown University, Washington, DC, United States.
Yoonseo Mok, Department of Epidemiology, University of Michigan, Ann Arbor, MI, United States.
Steve Cook, Department of Epidemiology, University of Michigan, Ann Arbor, MI, United States.
Rafael Meza, Rutgers Cancer Institute, Rutgers University, New Bunswick, NJ, United States; Population Health Sciences, BC Cancer Research Institute, Vancouver, British Columbia, Canada; School of Population and Public Health, University of British Columbia, Vancouver, British Columbia, Canada.
Author contribution
Kenneth Michael Cummings (Conceptualization; Formal analysis; Funding acquisition; Methodology; Writing—original draft; Writing—review & editing), Avery Roberson (Formal analysis; Writing—review & editing), Colin W. Randol (Formal analysis; Writing—review & editing), Tracy T. Smith (Writing—review & editing), David T. Levy (Funding acquisition; Writing—review & editing), Yoonseo Mok (Data curation; Formal analysis; Methodology; Writing—review & editing), Steve Cook (Writing—review & editing), and Rafael Meza (Conceptualization; Data curation; Formal analysis; Funding acquisition; Methodology; Writing—review & editing)
Supplementary material
Supplementary material is available at JNCI Cancer Spectrum online.
Funding
This research was supported by grant U54CA229974 from the United States National Cancer Institute (NCI) and United States Food and Drug Administration’s Center for Tobacco (FDA); and P01CA200512 from the NCI.
Conflicts of interest
Y.M., D.T.L., and R.M. reported grant funding from the NCI and FDA during the conduct of the study. K.M.C. reported funding from the NCI. K.M.C. also has been a paid expert witness in litigation against the cigarette industry. All others have no conflicts of interest to declare.
Data availability
The harmonized NHIS data from 2015 to 2023 are available from IPUMS NHIS.22 The data used in the analyses presented in the paper are available in the article and in its online supplementary material.
References
- 1. Institute of Medicine. Clearing the Smoke: Assessing the Science Base for Tobacco Harm Reduction. The National Academies Press; 2001. 10.17226/10029. [DOI] [PubMed] [Google Scholar]
- 2. National Academies of Sciences, Engineering, and Medicine. Public Health Consequences of E-Cigarettes. The National Academies Press; 2018. 10.17226/24952. [DOI] [PubMed] [Google Scholar]
- 3. Gottlieb S, Zeller M. A nicotine-focused framework for public health. N Engl J Med. 2017;377:1111-1114. [DOI] [PubMed] [Google Scholar]
- 4. US Department of Health and Human Services. The Health Consequences of Smoking: 50 Years of Progress: A Report of the Surgeon General. US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health; 2014.
- 5. Meza R, Cao P, Jeon J, Warner KE, Levy DT. Trends in US adult smoking prevalence, 2011 to 2022. JAMA Health Forum. 2023;4:e234213. 10.1001/jamahealthforum.2023.4213 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Blondino CT, Quezada FM, Hall C, Williams RT, Clifford JS. Generational differences in tobacco use patterns: results from wave 6 (2021) of the Population Assessment of Tobacco and Health Study (United States). J Stud Alcohol Drugs. 2026;87:398-408. 10.15288/jsad.25-00115 [DOI] [PubMed] [Google Scholar]
- 7. Mok Y, Cummings KM, Randol CW, et al. Joinpoint trend analysis of prevalence of combustible and non-combustible tobacco product use by adults in the United States, using cross-sectional data from NHIS 2015-2023. Tob Induc Dis. 2026;24:5. 10.18332/tid/213343 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Balfour DJ, Benowitz NL, Colby SM, et al. Balancing consideration of the risks and benefits of e-cigarettes. Am J Public Health. 2021;111:1661-1672. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Bozier J, Chivers EK, Chapman DG, et al. The evolving landscape of e-cigarettes: a systematic review of recent evidence. Chest. 2020;157:1362-1390. [DOI] [PubMed] [Google Scholar]
- 10. Mistry R, Bondarenko I, Jeon J, et al. Latent transition analysis of use frequencies for multiple nicotine and tobacco products among US adults. Addict Behav. 2025;161:108217. 10.1016/j.addbeh.2024.108217 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Brouwer AF, Jeon J, Jimenez-Mendoza E, et al. Changing patterns of cigarette and ENDS transitions in the USA: a multistate transition analysis of adults in the PATH study in 2017-2019 vs 2019-2021. Tob Control. 2025;34:783-790. 10.1136/tc-2023-058453 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Selya A, Kim S, Shiffman S, Goldenson N. Repeated point-prevalence of switching away from smoking after electronic nicotine delivery systems (ENDS) purchase. Harm Reduct J. 2025;22:71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Dai H, Benowitz NL, Achutan C, Farazi PA, Degarege A, Khan AS. Exposure to toxicants associated with use and transitions between cigarettes, e-cigarettes, and no tobacco. JAMA Netw Open. 2022;5:e2147891. 10.1001/jamanetworkopen.2021.47891 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Goniewicz ML, Smith DM, Edwards KC, et al. Comparison of nicotine and toxicant exposure in users of electronic cigarettes and combustible cigarettes. JAMA Netw Open. 2018;1:e185937. 10.1001/jamanetworkopen.2018.5937 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Chang CM, Cheng YC, Cho TM, et al. Biomarkers of potential harm: summary of an FDA-sponsored public workshop. Nicotine Tob Res. 2019;21:3-13. 10.1093/ntr/ntx273 [DOI] [PubMed] [Google Scholar]
- 16. Chang JT, Vivar JC, Tam J, et al. Biomarkers of potential harm among adult cigarette and smokeless tobacco users in the PATH study wave 1 (2013-2014): a cross-sectional analysis. Cancer Epidemiol Biomarkers Prev. 2021;30:1320-1327. 10.1158/1055-9965.EPI-20-1544 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Chang CM, Thakur S, Montes de Oca R, et al. Assessing the relationship between biomarkers of exposure and biomarkers of potential harm: PATH study wave 1 (2013 to 2014). Cancer Epidemiol Biomarkers Prev. 2024;33:1083-1090. 10.1158/1055-9965.EPI-23-1471 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Czoli CD, Fong GT, Goniewicz ML, Hammond D. Biomarkers of exposure among "dual users" of tobacco cigarettes and electronic cigarettes in Canada. Nicotine Tob Res. 2019;21:1259-1266. 10.1093/ntr/nty174 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Hartmann-Boyce J, Butler AR, Theodoulou A, et al. Biomarkers of potential harm in people switching from smoking tobacco to exclusive e-cigarette use, dual use or abstinence: secondary analysis of Cochrane systematic review of trials of e-cigarettes for smoking cessation. Addiction. 2023;118:539-545. 10.1111/add.16063 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Coleman SRM, Piper ME, Byron MJ, Bold KW. Dual use of combustible cigarettes and E-cigarettes: a narrative review of current evidence. Curr Addict Rep. 2022;9:353-362. 10.1007/s40429-022-00448-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Lisha NE, Vora M, Chaffee BW, Cheng J. Identifying and understanding use trajectories of cigarettes and E-cigarettes. Subst Use Misuse. 2025;60:2201-2213. 10.1080/10826084.2025.2537138 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Blewett LA, Rivera Drew JA, King ML, et al. IPUMS Health Surveys: National Health Interview Survey, Version 7.4 [dataset]. IPUMS; 2024.
- 23. Delnevo CD, Gundersen DA, Manderski MTB, Giovenco DP, Giovino GA. Importance of survey design for studying the epidemiology of emerging tobacco product use among youth. Am J Epidemiol. 2017;186:405-410. 10.1093/aje/kwx031 [DOI] [PubMed] [Google Scholar]
- 24. Surveillance Research Program US National Cancer Institute. Joinpoint trend analysis software. 2020. Accessed March 5, 2024. https://surveillance.cancer.gov/joinpoint
- 25. Kim HJ, Chen HS, Midthune D, et al. Data-driven choice of a model selection method in joinpoint regression. J Appl Stat. 2023;50:1992-2013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Kim HJ, Chen HS, Byrne J, et al. Twenty years since Joinpoint 1.0: two major enhancements, their justification, and impact. Stat Med. 2022;41:3102-3130. [DOI] [PubMed] [Google Scholar]
- 27. Kleykamp BA, Heishman SJ. The older smoker. JAMA. 2011;306:876-877. 10.1001/jama.2011.1221 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Williamson TJ, Riley KE, Carter-Harris L, Ostroff JS. Changing the language of how we measure and report smoking status: implications for reducing stigma, restoring dignity, and improving the precision of scientific communication. Nicotine Tob Res. 2020;22:2280-2282. 10.1093/ntr/ntaa141 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Borland R, Murray K, Gravely S, et al. A new classification system for describing concurrent use of nicotine vaping products alongside cigarettes (so-called 'dual use’): findings from the ITC-4 Country Smoking and Vaping Wave 1 Survey. Addiction. 2019;114(Suppl 1):24-34. 10.1111/add.14570 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Rest EC, Mermelstein RJ, Hedeker D. Nicotine dependence in dual users of cigarettes and E-cigarettes: common and distinct elements. Nicotine Tob Res. 2021;23:662-668. 10.1093/ntr/ntaa217 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. McNeill A, Driezen P, Hitchman SC, Cummings KM, Fong GT, Borland R. Indicators of cigarette smoking dependence and relapse in former smokers who vape compared with those who do not: findings from the 2016 International Tobacco Control Four Country Smoking and Vaping Survey. Addiction. 2019;114(Suppl 1):49-60. 10.1111/add.14722 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Halliday DM, Song AV, Nguyen N. E-cigarette and cannabis use among current and recently quit smokers: co-use and co-cessation. Addict Behav Rep. 2025;21:100611. 10.1016/j.abrep.2025.100611 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Dai HD, Leventhal AM. Prevalence of nicotine pouch use among US adults. JAMA. 2024;332:755-757. 10.1001/jama.2024.10686 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The harmonized NHIS data from 2015 to 2023 are available from IPUMS NHIS.22 The data used in the analyses presented in the paper are available in the article and in its online supplementary material.