Association Between Cigarette Smoking Frequency and Tobacco Use Disorder in U.S. Adults

INTRODUCTION

Cigarette smoking remains a major cause of premature death and nicotine dependence is a major barrier to smoking cessation.^1,2 The DSM-5 is the primary guide to psychiatric diagnoses in the U.S. The 5th revision, released in 2013, replaced the term “nicotine dependence” and instead assigns the diagnosis “Tobacco Use Disorder” (TUD) to individuals experiencing clinically significant impairment, as indicated by meeting at least 2 of 11 specified criteria.³ Clinical practice guidelines encourage clinicians to routinely assess and treat tobacco use and dependence.⁴ Patient’s self-reported smoking rate is often used to guide clinical decision making, but surprisingly no previous report has described the proportion of cigarette smokers who meet DSM-5 TUD criteria across the full range of cigarette use patterns in a representative sample of the U.S. population. Lighter smoking is frequently perceived as less harmful. Consequently, such smokers may be less likely to be identified by medical providers and less likely to receive appropriate treatment.⁵ Information on the true prevalence of TUD in this population can directly inform clinical practice.

METHODS

The authors analyzed data from the National Epidemiological Study of Alcohol and Related Conditions–III, a nationally representative study of the non-institutionalized adult civilian population in the U.S. conducted from 2012 to 2013. Analyses were conducted in 2020. A total of 36,309 individuals (60.1% response rate) participated. The present analysis included individuals who reported past-year cigarette use, no use of other tobacco products, and had smoking rate information available (n=6,793). In-person interviews were conducted to assess the presence and severity of DSM-5 TUD among smokers. A detailed description of study procedures and interviews is available elsewhere.^6–8 Details on individual symptom coding and the analytic approach are available in the Appendix. This project was deemed exempt by the Duke University Health System IRB.

RESULTS

Figure 1 illustrates the proportion of cigarette smokers at each level of cigarette consumption who met TUD criteria, and the mean number of criteria (of 11) met at each level of cigarette consumption. A clear relationship between the number of cigarettes smoked on a typical smoking day and the probability of meeting TUD criteria was also evident for non-daily smokers. Even among those who smoked less than weekly, 18.3% (95% CI=13.7, 24.2) of those who typically smoked 1–2 cigarettes on a smoking day met TUD criteria, whereas 44.6% (95% CI=36.8, 52.6) of those who smoked ≥3 cigarettes on a smoking day met TUD criteria. The Appendix provides additional details.

Figure 1.

Relationship of cigarette smoking frequency with tobacco use disorder and mean symptoms endorsed.

Notes: Bars represent 95% CIs.

TUD, tobacco use disorder.

DISCUSSION

Results reveal a clear relationship between frequency of cigarette smoking and the proportion of cigarette smokers meeting DSM-5 TUD criteria, which becomes relatively flat above 10 cigarettes per day (CPD), at around 90%. Notably, almost two thirds of those smoking only 1–4 CPD met TUD criteria, as did the majority (64.1%) of those smoking 3–6 days per week and a substantial minority (26.3%) of individuals who smoke less than once per week. DSM-5 states that TUD is considered of “moderate” severity when ≥4 criteria are met, and these data show this is typical at ≥10 CPD. The proportion of smokers meeting diagnostic criteria at specific levels of cigarette consumption appears greater than in prior studies using DSM-4 criteria for nicotine dependence.^9,10 This may be related to numerous factors, including changes in diagnostic criteria to encompass a dimensional approach that includes mild TUD (meeting 2–3 of 11 criteria), as well as the substantial (42%) increase in average nicotine intake per cigarette in recent decades.¹¹ The present study highlights the high prevalence of TUD even among those considered light smokers and potential need for treatment within this population. Previous research has found that low-rate smokers report greater motivation to quit and are more likely than daily smokers to make a quit attempt.^12,13 However, it is unclear the extent to which existing interventions are effective for light smokers. Continued efforts to identify optimal cessation approaches for this population remain an important direction for future research. Strengths of this study include the large representative sample spanning the full range of smoking patterns, and the use of a validated diagnostic assessment.

Limitations

Limitations include reliance on self-report and the exclusion of individuals who use multiple tobacco products, who represent a rapidly growing segment of the population of smokers.

CONCLUSIONS

The overwhelming majority (85.0%) of daily cigarette smokers and a sizable minority of non-daily smokers (44.0%) meet DSM-5 diagnostic criteria for TUD. Clinicians should ask about all smoking behavior, including non-daily smoking, as such smokers may still require treatment to successfully quit smoking.