Patterns and Characteristics of Nicotine Dependence Among Adults With Cigarette Use in the US, 2006-2019

Key Points Question Does the national prevalence of nicotine dependence vary by year, age, psychiatric comorbidities (substance use disorder and/or depression), and sociodemographic characteristics? Findings In this cross-sectional study of 152 354 US community-dwelling adults with past-month cigarette use, the adjusted prevalence of nicotine dependence decreased from 59.52% in 2006 to 56.00% in 2019 among adults with past-month cigarette smoking and among each age group, except for ages 18 to 25 years. Adults 50 years and older (especially those with substance use disorder and/or depression) had the highest prevalence of nicotine dependence. Meaning These findings suggest that evidence-based tobacco cessation strategies tailored to age and psychiatric comorbidities are needed.


Introduction
Nicotine dependence increases the risk of smoking persistence and is the leading preventable cause of morbidity and death. 1 The prevalence of cigarette smoking has declined in the US over the past decades among adults with and without psychiatric conditions (eg, major depressive episode [MDE]   and substance use disorder [SUD]). 2,3A nicotine hardening hypothesis has emerged.This hypothesis proposes that declines in nicotine use resulting from population-level tobacco control measures leave a higher proportion of people with intractable nicotine dependence and consumption over time. 4wever, findings on the hardening hypothesis from existing literature [4][5][6] are inconsistent.
Based on nationally representative data collected from 2001 to 2002 and 2012 to 2013, Grant et al 4 reported increases in nicotine dependence prevalence among US adults with nicotine use, supporting the hardening hypothesis.Methodological differences between the 2001 to 2002 and 2012 to 2013 data collections may limit comparability for the 2 time periods.Another study 5 based on representative data from 18 European countries provided empirical evidence against this theory.One study 6 reviewed 26 population-based studies and did not find that conversion from current to former smoking, number of quit attempts, or success on a given quit attempt decreased over time, suggesting hardening may not be occurring in noninstitutionalized civilian populations with cigarette use.Moreover, evidence regarding whether and how nicotine dependence varies by age is inconsistent.Studies have reported that a significant decrease in nicotinic activation in the hippocampus (a brain region involved in addiction) occurs around age 40 years, 7,8 and reduced nicotinic activation is associated with reduced addictive potential of nicotinic agonists. 9Another study 10 based on nationally representative data from 1991 to 1993 reported that US adults 50 years and older who smoke cigarettes had the highest consumption yet experienced the lowest prevalence of nicotine dependence.One study 11 suggested that this older age group may be more responsive to nonpharmacological cessation programs because of their reduced nicotine dependence.However, the study from 18 European countries 5 found that high nicotine dependence was associated with older age.Another recent study 12 based on nationally representative data from China found that the prevalence of nicotine dependence among adults with cigarette use increased with age for men but peaked at age 40 years for women.In addition, although nicotine dependence is associated with lower socioeconomic status 1 and psychiatric disorders, 13 little is known about whether and how associations of sociodemographic characteristics and psychiatric conditions with nicotine dependence vary by age.
Given continuing declines in smoking and because the hardening hypothesis could play a role in US tobacco cessation strategies, this study assessed patterns in nicotine dependence among US adults with cigarette use based on nationally representative data from 2006 through 2019.Among adults with cigarette use, patterns and characteristics of nicotine dependence were examined to understand whether and how associations of sociodemographic characteristics and psychiatric conditions with nicotine dependence varied by age.

JAMA Network Open | Substance Use and Addiction
Nicotine Dependence Among Adults With Cigarette Use in the US, 2006-2019

Data Sources
Data were from 152 354 US community-dwelling individuals 18 years and older who participated in the 2006-2019 National Surveys on Drug Use and Health (NSDUH); the NSDUH provides nationally representative data on cigarette smoking, nicotine dependence, MDE, and SUD among US noninstitutionalized civilian adult populations. 14,15The NSDUH data collection protocol was approved by the institutional review board of Research Triangle Institute International.Each participant provided verbal informed consent. 14,15This cross-sectional study used publicly available deidentified data from the NSDUH and was exempt from review and the requirement for informed consent per the institutional review board of the National Institutes of Health.The study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.
For the NSDUH, data were collected by interviewers through personal visits to households and noninstitutional group facilities.Audio computer-assisted self-administered interviewing was used, providing respondents with a private and confidential way to record answers.From 2006 to 2019, the annual mean weighted screening response rate was 82.9%, and the annual mean weighted interview response rate was 71.4%.Data collected for this cross-sectional study were analyzed from January 15 to February 15, 2023.

Measures
Among NSDUH respondents who reported past-month cigarette smoking, past-month nicotine dependence was defined as meeting criteria from either the Nicotine Dependence Syndrome Scale or the Fagerström Test of Nicotine Dependence. 13Using diagnostic criteria specified in the Diagnostic and Statistical Manual of Mental Disorders (Fourth Edition), Text Revision (DSM-IV-TR), 2,14-17 the NSDUH estimated the prevalence of past year-specific SUD, and respondents were classified as having past-year SUD if they had any of 4 specific SUDs (eg, alcohol use disorder, cannabis use disorder, cocaine use disorder, and heroin use disorder). 25][16][17] These measures of SUDs and MDE have been found to have good validity and reliability. 18,19In addition, the NSDUH asked respondents about sociodemographic characteristics (eg, age, sex, race and ethnicity [including Hispanic; non-Hispanic Black; non-Hispanic White; and other non-Hispance race, including American Indian or Alaska Native, Asian, Native Hawaiian or other Pacific Islander, and more than 1 race]), educational attainment, employment status, family income, marital status, health insurance status, and metropolitan statistical area [MSA] status).[16]

Statistical Analysis
Bivariable and multivariable logistic regression analyses were conducted to understand factors associated with past-month nicotine dependence among adults with past-month cigarette smoking.
Model-adjusted prevalence (predicted marginal proportion) and differences in model-adjusted prevalence (differences in predicted marginal proportions) were estimated.Multicollinearity and potential interaction effects were examined.Although multicollinearity was not found in the final pooled model, significant interaction effects were identified between age and selected covariates.Thus, stratified multivariable logistic regression models by age were conducted.
All of these analyses used SUDAAN software, release 11.0.3 (RTI International), 20  valid even if the Joinpoint models suggest changes in patterns during a given study period. 22For each analysis, 2-tailed P < .05 was considered statistically significant.

Overall Patterns and Characteristics of Nicotine Dependence
Among 152 354 community-dwelling adults with past-month cigarette use, 54.1% were male, 45.9%
Nicotine dependence prevalence did not differ by employment, health insurance, or marital status among those 50 years and older with cigarette use.However, prevalence was 18% lower among those aged 18 to 25 years with part-time employment than their counterparts with full-time employment (ARR, 0.82; 95% CI, 0.79-0.84)(Table 2).Among adults aged 18 to 49 years, prevalence was higher among those who were unemployed vs those who had full-time employment, ranging from 5% higher for those aged 35 to 49 years (ARR, 1.05; 95% CI, 1.00-1.09)to 17% higher for those aged 26 to 34 years (ARR, 1.17; 95% CI, 1.11-1.23).

Nicotine Dependence Prevalence by Age, Educational Attainment, and MSA
Across age groups, nicotine dependence prevalence was higher among those without a college degree vs those with a college degree, ranging from 20% higher for those 50 years and older with some college education (ARR, 1.20; 95% CI, 1.12-1.27)to 98% higher for those aged 26 to 34 years (ARR, 1.98; 95% CI, 1.86-2.13)and 136% higher for those aged 18 to 25 years with less than a high school diploma (ARR, 2.36; 95% CI, 2.20-2.53)(Table 2).Compared with adults 50 years and older, differences in nicotine dependence prevalence between those without a high school diploma vs those with a college degree were higher among those aged 18  Across age groups, nicotine dependence prevalence was lower among those residing in large MSAs than those in non-MSAs, ranging from 3% lower for those 50 years and older (ARR, 0.97; 95% CI, 0.94-0.99) to 11% lower for those aged 26 to 34 years (ARR, 0.89; 95% CI, 0.86-0.92)(Table 2).

Nicotine Dependence Prevalence by Age, SUD, and MDE
Overall, nicotine dependence was higher among adults with cigarette use who also had SUD and/or MDE than among those without these comorbid conditions, but the pattern of associations varied according to age group (Table 1 and Table 2; Figure 2).For example, the differences in nicotine dependence prevalence between those with co-occurring SUD and MDE and those with neither condition were more than 2 times larger for adults 50 years and older vs those aged 18 to 49 years percentage point difference [57.76% vs 49.48%]; P < .001)(Table 1).In all age groups of adults with cigarette use who had neither SUD nor MDE, nicotine dependence prevalence was lower than among adults with cigarette use who had SUD but not MDE, ranging from 6% lower for those aged 35 to 49 years (ARR, 0.94; 95% CI, 0.91-0.98) to 15% lower for those 50 years and older (ARR, 0.85; 95% CI, 0.81-0.88)(Table 2).However, also compared with those with SUD but without MDE, nicotine dependence prevalence was higher only among those aged 18 to 25 years (ARR, 1.09; 95% CI, 1.04-1.14),those aged 26 to 34 years (ARR, 1.07; 95% CI, 1.01-1.14),and those 50 years and older (ARR, 1.09; 95% CI, 1.02-1.17)with co-occurring SUD and MDE.Prevalence was also higher among those aged 26 to 34 years (ARR, 1.10; 95% CI, 1.04-1.16)and those aged 35 to 49 years (ARR, 1.11; 95% CI, 1.06-1.16)with MDE but without SUD compared with those with SUD but without MDE.

Figure 1 .
Figure 1.Patterns in Adjusted Past-Month Prevalence of Nicotine Dependence Among US Adults With Past-Month Cigarette Smoking by Age

Table 2
Nicotine Dependence Among Adults With Cigarette Use in theUS, 2006-2019 ). Prevalence was lower among minority racial and ethnic adults than non-Hispanic White adults across examined age groups; difference in prevalence ranged from 9% lower for non-Hispanic individuals of other race who were 50 years and older (ARR, 0.91; 95% CI, 0.85-0.97) to 52% lower for Hispanic individuals aged 18 to 25 years (ARR, 0.48; 95% CI, 0.46-0.51).However, compared with those 50 years and older, differences in nicotine dependence prevalence betweenJAMA Network Open | Substance Use and Addiction

Table 1 .
Adjusted Past-Month Prevalence of Nicotine Dependence Among Adults With Past-Month Cigarette Smoking in the US a .001),marital status and age (P < .001),metropolitan statistical area and age (P = .01),and psychiatric condition and age (P < .001).Hispanic American Indian or Alaska Native, non-Hispanic Asian, non-Hispanic Native Hawaiian or other Pacific Islander, and non-Hispanic more than 1 race.Race and ethnicity were determined according to National Survey on Drug Use and Health respondents' self-classification of racial and ethnic origin and identification based on US Census Bureau classifications.
e Some college Abbreviations: MDE, major depressive episode; NA, not applicable; SUD, substance use disorder (includes alcohol use disorder, cannabis use disorder, cocaine use disorder, and heroin use disorder).a Results of the final pooled and stratified multivariable logistic regression models with these variables.All data were obtained from the 2006-2019 National Surveys on Drug Use and Health.b The final sample size was 152 354 rather than 154 394 due to the unknown MDE status of 2040 respondents (1.3%).c Also includes the following interaction effects: race and ethnicity and age (P < .001),employment status and age (P < .001),family income and age (P = .007),educational attainment and age (P < .001),health insurance status and age (P < d Reference group.e Estimates were significantly different (P < .05)from the corresponding reference group.f Includes non-

Table 2 .
Characteristics Associated With Past-Month Dependence Among Adults With Past-Month Cigarette Smoking in the US a Results of the final pooled and stratified multivariable logistic regression models with these variables.All data were obtained from the 2006 to 2019 National Surveys on Drug Use and Health..001),marital status and age (P < .001),metropolitan statistical area and age (P = .01),and psychiatric condition and age (P < .001).Hispanic American Indian or Alaska Native, non-Hispanic Asian, non-Hispanic Native Hawaiian or other Pacific Islander, and non-Hispanic more than 1 race.Race and ethnicity were determined according to National Survey on Drug Use and Health respondents' self-classification of racial and ethnic origin and identification based on US Census Bureau classifications.non-Hispanic White vs non-Hispanic Black groups smaller for those aged 18 to 25 years (difference, 7.55 percentage points [48.29% vs 40.74%] vs 14.13 percentage points [70.50% vs 56.37%]; P < .001)and those aged 26 to 34 years (difference, 7.23 percentage points [57.62% vs 50.39%] vs 14.13 percentage points [70.50% vs 56.37%]; P < .001)(Table 1).Compared with those 50 years and older, differences in prevalence between non-Hispanic White and Hispanic groups were larger for those aged 26 to years (difference, 29.14 percentage points [57.62% vs 28.48%] vs 23.81 JAMA Network Open.2023;6(6):e2319602.doi:10.1001/jamanetworkopen.2023.19602(Reprinted) June 23, 2023 7/15 Downloaded From: https://jamanetwork.com/ on 11/03/2023 2. Characteristics Associated With Past-Month Nicotine Dependence Among Adults With Past-Month Cigarette Smoking in the US a (continued) d Abbreviations: MDE, major depressive episode; NA, not applicable; SUD, substance use disorder (includes alcohol use disorder, cannabis use disorder, cocaine use disorder, and heroin use disorder).a c Also includes the following interaction effects: race and ethnicity and age (P < .001),employment status and age (P < .001),family income and age (P = .007),educational attainment and age (P < .001),health insurance status and age (P < d Estimates were significantly different (P < .05)from the corresponding reference group.e Includes non- to 25 years (difference, 31.58 percentage points [54.79% vs 23.21%] vs 19.40 percentage points [73.68% vs 54.28%]; P < .001)and those aged 26 to 34 years (difference, 31.55 percentage points [63.71% vs 32.16%] vs 19.40 percentage points; P < .001)(Table 1).