___________________

¹ Note that quotations are used at the first occurrence of the term “premium” in each chapter, as there is no formally agreed upon definition of what constitutes a premium cigar, and different entities might use this term differently. See Chapter 1 for more information. In addition, when the terms “cigar(s)” or “cigars in general” are used in this report, they refer to all cigar types (filtered cigars, little cigars, cigarillos, and large/traditional cigars [which include premium cigars]). When discussing a specific cigar type, the type is noted in text.

cigars. The toxicants generated by combustion of tobacco include oxidizing chemicals, carcinogens (such as nitrosamines and polycyclic hydrocarbons), carbonyls (such as acrolein, formaldehyde, and acetaldehyde), carbon monoxide (CO), metals, and particulates. When these substances are inhaled, oxidant stress, systemic inflammation, endothelial dysfunction, DNA damage, hypercoagulability, and changes in microbial populations occur, which lead to organ dysfunction and disease (HHS, 2010). While inhalation is necessary to deliver toxicants to the heart, lungs, and other body organs, taking tobacco smoke into the mouth without inhalation exposes the mouth, pharynx, and esophagus to various toxicants. Thus, upper respiratory tract and esophageal disease can occur even in tobacco users who do not inhale, which is typical of some premium cigar smokers.

While products of combustion are thought to be responsible for most of the disease caused by smoking, nicotine may also contribute. Nicotine comes from the tobacco itself, and its absorption does not require combustion or even inhalation. Nicotine is a weak base, and, in the presence of an alkaline pH, is readily absorbed across mucous membranes, such as the mouth. Users of smokeless tobacco products absorb, on average, similar amounts of nicotine each day as do cigarette users (Piano et al., 2010). While fewer data are available on cigar users who do not inhale, the potential for substantial nicotine exposure is clear (see Chapter 2). In contrast to cigarettes, where the wrapper is paper, cigar wrappers contain tobacco, so nicotine can be absorbed orally through direct contact with the tobacco leaf, as well as through the smoke.

The most important harmful effect of nicotine is sustained use of combusted tobacco by causing addiction (Benowitz, 2010); evidence of addiction in premium cigar users is discussed later in this chapter. Other effects of concern include contributions to CVD, increased risk of diabetes and pro-atherogenic lipid profiles, reproductive toxicity, including low birth weight and effects on fetal neurodevelopment, and possible adverse effects on adolescent brain maturation (Benowitz and Burbank, 2016; HHS, 2010). Nicotine also releases catecholamines, which cause constriction of blood vessels, which in turn may promote oral pathology or result in impaired wound healing after surgical procedures.

Other tobacco and substance use behaviors also need to be considered in assessing potential adverse effects of premium cigar use. Former or concurrent cigarette or small cigar users are more likely to inhale more intensively compared to users who have smoked exclusively large and premium cigars (see Chapter 2). All tobacco products, including premium cigars, are commonly used in conjunction with drinking alcoholic beverages (see Chapter 3). Alcohol and smoking act synergistically to increase the risk of head and neck and esophageal cancer.

ucts, exerting its effects by stimulation of nicotine acetylcholine receptors. In its unprotonated state, it readily crosses cell membranes and enters the body. When inhaled, unprotonated nicotine is more volatile and acts on nicotinic cholinergic receptors in the mouth and throat, producing sensations of irritation and harshness (Benowitz et al., 2021; Leventhal et al., 2021). Since cigar smoke is often more alkaline than smoke from cigarettes or small cigars, more nicotine in the smoke is in the unprotonated state, so the smoke is more irritating and difficult to inhale (see Chapter 2). For this reason, the pH of cigar smoke is a critical factor influencing whether or how deeply a user inhales, and the expression of nicotine’s effects. Other established properties of nicotine include acute toxicity at high doses and negative effects on maternal and fetal health (England et al., 2017). The negative effects of nicotine withdrawal on cognitive function have been established. It is also associated with dysphoric mood, including anxiety and depression. Relief of these symptoms is rewarding (termed “negative reinforcement”) and contributes to nicotine addiction. Nicotine exposure during adolescence causes long-term structural and functional changes to the brain in rodents and might also do so in human youth (HHS, 2014). Some evidence exists for abuse potential of the minor tobacco alkaloids nornicotine and anabasine, which could play a role in abuse potential of tobacco products, including premium cigars (Hoffman and Evans, 2013). Some evidence also suggests the endogenous nitrosation of nornicotine in the human body, leading to formation of the carcinogen NNN (Knezevich et al., 2013; Stepanov et al., 2009).

Considerable amounts of CO (e.g., 97 mg/cigar in the Macanudo premium cigar brand) are present in premium cigar smoke (NCI, 1998). CO binds rapidly with hemoglobin in the blood, diminishing its oxygen-carrying capacity and potentially leading to a number of negative health effects, particularly in people with underlying cardiovascular or pulmonary disease. Ammonia causes a burning sensation in the eyes, nose, throat, and respiratory tract, and hydrogen cyanide is a known poison.

Acetaldehyde has some addiction potential and binds to DNA (Balbo et al., 2012; FDA, 2012). As the major initial metabolite of ethanol, most of its potentially toxic, genotoxic, and carcinogenic effects are associated with alcohol consumption (IARC, 2012c). Acrolein (55–60 µg/g cigar tobacco smoked) is one of the most irritating and toxic compounds in tobacco smoke (Haussmann, 2012; NCI, 1998). It reacts with DNA to form well-characterized adducts (Paiano et al., 2020) and is considered probably carcinogenic to humans by the International Agency for Research on Cancer (IARC) (IARC Monographs Vol 128 group, 2021). Benzene is a known human carcinogen causing acute myeloid leukemia/acute nonlymphocytic leukemia (IARC, 2012e), and 1,3-butadiene is considered “carcinogenic to humans” by IARC, causing cancer of the hematolym-

phatic organs (IARC, 2012e). IARC also considers ethylene oxide and formaldehyde carcinogenic to humans. The former is based on evidence from studies in laboratory animals and compelling evidence from genotoxicity studies in humans (IARC, 2012c). Formaldehyde is an accepted cause of cancer of the nasopharynx and of leukemia (IARC, 2012d).

Tobacco-specific nitrosamines are among the most well-characterized carcinogens in unburned tobacco and tobacco smoke, with consistently problematic levels in both, as summarized in Chapter 2. NNK causes tumors of the lung in all species tested, including rats, hamsters, ferrets, and multiple strains of mice, independent of the route of administration (Hecht, 1998). The lowest total dose of NNK shown to induce lung tumors in rats was 1.8 mg/kg body weight. The carcinogenicity of NNN has also been established in multiple species, including various strains of rats and mice and in Syrian golden hamsters and mink (Hecht, 1998). In one study, chronic oral administration of (S)-NNN (the major form in tobacco) at a dose of 14 ppm in drinking water induced 89 benign and malignant oral cavity tumors and 122 esophageal tumors in a group of 20 rats (Balbo et al., 2013). IARC considers NNK and NNN, which always occur together in tobacco and tobacco smoke and are present in all tobacco products, carcinogenic to humans (IARC, 2012b).

Polycyclic aromatic hydrocarbons (PAH) represent a well-established class of carcinogens that are formed by incomplete combustion of organic matter, including tobacco. They are widespread environmental contaminants also found in air, water, soils, sediments, and broiled foods. Multiple PAH are present in tobacco and tobacco smoke. One study tentatively identified more than 500 different PAH in tobacco smoke condensate fractions (Snook et al., 1978). Benzo[a]pyrene (BaP) is the most extensively investigated, and many of its properties are illustrative of the class (IARC, 2012e). BaP induces tumors in laboratory animals by various routes of administration; is a complete carcinogen (affects tumor cells in all stages of development) and tumor initiator when applied to mouse skin; and induces tumors at the injection site and sometimes lungs in mice when administered by subcutaneous injection. Many other PAH have similar carcinogenic properties as those of BaP, which IARC considers carcinogenic to humans (IARC, 2012e).

Aromatic amines, such as 4-aminobiphenyl and 2-naphthylamine, are recognized human bladder carcinogens in tobacco smoke (IARC, 2012e). Heterocyclic aromatic amines are a broad class of well-established carcinogens formed during high-temperature combustion processes involved in food preparation as well as in cigarette and cigar smoking (Bellamri et al., 2021).

Phenols, such as catechol, occur in relatively high concentrations in cigarette smoke. Catechol is a co-carcinogen, enhancing the tumorigenic

activity of PAH in mouse skin models, and can be involved in oxidative damage by tobacco smoke. Phenols, while generally not carcinogenic themselves, may enhance the activity of carcinogens in smoke (Hecht, 2011).

Metals are among the HPHCs in tobacco and tobacco smoke; some of these, including lead, cadmium, and nickel, have been identified in cigar smoke (NCI, 1998). Prolonged exposure to lead may have a variety of health effects, including high blood pressure, heart disease, and kidney disease (CDC, 2018). Cadmium and nickel are considered carcinogenic to humans by IARC, causing lung cancer as well as some other cancer types (IARC, 2012a).

Summary and Conclusions

FDA has established a list of HPHCs in tobacco products and tobacco smoke. This list includes tobacco alkaloids such as nicotine, volatile inorganic compounds such as CO, volatile organic compounds such as the human carcinogen benzene, carcinogenic tobacco-specific nitrosamines and other nitrosamines, carcinogenic PAH, carcinogenic aromatic amines and heterocyclic aromatic amines, toxic phenols, carcinogenic metals, and miscellaneous other toxic and carcinogenic compounds. Cigar smoke, including premium cigar smoke, contains many HPHCs capable of causing cancer and multiple other negative health effects. Smokers of premium cigars will be exposed to this toxic and carcinogenic mixture when they use these products. The level of exposure to specific HPHCs in premium cigar users will depend on how the cigars are smoked, including frequency of smoking and depth of inhalation. The overall exposure to HPHCs in daily users of traditional cigars appears to be similar to daily smokers of nonpremium cigars and daily smokers of cigarettes. Biomarker data showing equivalent exposure comparing daily traditional cigar and cigarette smokers are based on data from a combination of large and premium cigars, however based on biological plausibility, it is likely the results would apply to premium cigar only users as well.

Conclusions 5-1 and 5-2 are based on the known chemical characteristics of combustible tobacco products, including cigars, as well as biological mechanisms by which constituents of combustible tobacco products are processed (in animals and humans). While studies on cigars may include premium cigars, they do not distinguish premium from other cigar types. However, given the conclusive data on tobacco products including cigars in general, and the absence of any important threats to validity, the committee extrapolated these findings to premium cigars.

___________________

² Several studies cover more than one health outcome—study details are included at first occurrence in the chapter. The Chapter 5 Annex includes evidence tables for studies that review primary cigar smoking.

³ Some studies use the comparison group of “never-tobacco users” as opposed to “never-tobacco smokers,” as the reference group also excluded users of smokeless tobacco products.

mortality among secondary cigar smokers compared to never-smokers (NCI, 1998). One study (Ben-Shlomo et al., 1994), using data from 19,018 Whitehall Study participants (men aged 40–69 from the British Civil Service), observed lower mortality rates for current primary cigar smokers compared to never-smokers (age-adjusted mortality rate per 1,000 person-years was 5.04 for primary cigar smokers versus 10.50 for never-smokers). However, this study included relatively few deaths among primary cigar smokers (9 of the 4,496 total deaths were among primary cigar smokers).

Since these reviews, four additional prospective epidemiologic studies examining the association of cigar smoking and mortality have been published (Christensen et al., 2018; Inoue-Choi et al., 2019; Rodu and Plurphanswat, 2021; Thomson et al., 2020), including three of nationally representative longitudinal cohorts conducted in the United States (Christensen et al., 2018; Inoue-Choi et al., 2019; Rodu and Plurphanswat, 2021). The results are largely consistent with those included in the prior reviews and support the conclusion of an increased risk of mortality. Christensen et al. (2018) evaluated the association of cigarette, cigar, and pipe use with cause-specific mortality and other health outcomes in the National Longitudinal Mortality Study (NLMS; n = 357,420). The NLMS is a mortality follow-up of the Tobacco Use Supplement to the Current Population Survey starting in 1985 followed through 2011, including from 1,139 exclusive current and 2,398 exclusive former cigar users. The cigar use category included use of little cigars, cigarillos, or large cigars. Christensen et al. (2018) found increased all-cause mortality among current primary cigar smokers compared to never-tobacco users (hazard ratio [HR]: 1.20; 95 percent confidence interval [CI]: 1.03–1.38); this risk was attenuated among former primary cigar smokers (HR: 1.11; 95 percent CI: 0.99–1.25). Both Inoue-Choi et al. (2019) and Rodu and Plurphanswat (2021) assessed the risk of mortality associated with cigar use using data from the National Health Interview Survey (NHIS) Linked Mortality Files with follow-up through 2015. Inoue-Choi et al. used data from 1991, 1992, 1998, 2000, 2005, and 2010 to evaluate the association between exclusive cigarette, smokeless tobacco, pipe, and cigar use with overall and cause-specific mortality (Corrigendum, 2019; Inoue-Choi et al., 2019). The analysis included data from 165,335 adults at least 18 years of age, including 1,595 exclusive cigar users; however, type of cigar was not available. Rodu and Plurphanswat was restricted to men aged 40–79 (N = 52,710) and included data from NHIS 1987. Both studies found increased risk of mortality among current primary cigar smokers, though these findings were not statistically significant (incidence rate ratio (IRR) [95 percent CI] were 1.22 [0.93–1.60] for Inoue-Choi et al. and 1.02 [0.86–1.23] for Rodu and Plurphanswat). Thomson et al. (2020) studied 118,840 adults aged 30–69 in Cuba and found that compared to never-smokers, primary cigar smoking was associated with increased all-cause mortality (IRR: 1.27; 95 percent CI: 1.11–1.46).

1.22 (95 percent CI: 1.04–1.44), respectively, compared with never-tobacco users (Christensen et al., 2018).

Compared with never-tobacco users, current primary cigar smokers who smoked 1–2, 3–4, and ≥5 cigars/day had RRs of mortality of 1.02 (95 percent CI: 0.97–1.07), 1.08 (95 percent CI: 1.02–1.15), and 1.17 (95 percent CI: 1.10–1.24), respectively (NCI, 1998). Likewise, among secondary cigar smokers, smoking ≥3 cigars per day was associated with significantly increased risk of all-cause mortality (RRs, 95 percent CI: 1.17–1.18) (NCI, 1998).

Impact of Depth of Inhalation of Cigar Smoke on All-Cause Mortality

Only one study has examined the impact of inhalation depth on risk of mortality among cigar smokers (NCI, 1998). Among current primary cigar smokers in CPS-I, greater depth of inhalation was associated with significantly increased risk of all-cause mortality (IRRs [95 percent CI] were 1.04 [1.00–1.08], 1.19 [1.09–1.30], and 1.60 [1.38–1.84] for none, slight, and moderate-deep inhalation, respectively, compared to never-smokers) (NCI, 1998). Associations of inhalation depth and mortality were slightly reduced among secondary cigar smokers (IRRs [95 percent CI] were 1.04 [0.97–1.11], 1.16 [1.04–1.29], and 1.33 [1.16–1.51] for none, slight, and moderate-deep inhalation, respectively, compared to never-smokers) (NCI, 1998).

All-Cause Mortality Summary

Cigar use in general is associated with increased risk of all-cause mortality compared to never-tobacco use, with generally lower risk than that observed in cigarette smoking and higher risk than in pipe smoking. The risk for all-cause mortality with cigar smoking increases with daily cigar smoking, additional number of cigars smoked per day, and greater depth of inhalation.

Cigar use in studies of all-cause mortality do not distinguish premium from nonpremium cigars. That is, these studies may include premium cigars but also other large cigars, little cigars, or cigarillos. Studies that distinguish premium from nonpremium cigar use as well as studies that distinguish primary and secondary cigar smokers would better isolate the effects of premium cigar smoking. Information on frequency and intensity of cigar smoking, the depth of inhalation of cigar smoke, and the number of years smoking cigars would inform potential dose–response relationship and modifying factors. Most studies are conducted in predominately white male populations; the lack of studies that include women and racial and ethnic minority populations is a research gap.

___________________

⁴ Studies that primarily looked at pipe smokers or combined pipe and cigar smokers in the analysis were not included in this review (e.g., Wannamethee et al., 2005).

1998). There was generally a slight (and often nonsignificant) increased risk of coronary heart disease in primary and secondary cigar users, with variability across studies. There was a trend for increased risk based on the number of cigars smoked per day for primary cigar smokers (1–2 versus 3–4 versus 5+ per day) and for depth of inhalation (none versus slight versus moderate-deep); the risk in all cases was less than that of cigarette smoking.

Data from the CPS-I study found no increased risk of stroke death in either primary or secondary cigar smokers, regardless of the number of cigars smoked per day or the depth of inhalation. CPS-I found a significant increase in risk of aortic aneurysm deaths for both primary and secondary cigar smokers, with no clear dose–response or relationship to depth of inhalation.

Chang et al. (2015) reviewed coronary heart disease mortality. This analysis included published data from CPS-I and CPS-II. The authors overall found a slight and generally nonsignificant increased risk in cigar-only smokers, with the exception of CPS-II, in which those age 30–74 who inhaled moderately deeply or had smoked for 25 or more years had a significant HR of approximately 1.4.

Several studies not included in previous reviews, or that are particularly informative (e.g., large sample size) are summarized here. Wald and Watt (1997) used a prospective cohort study of British men 35–64 years old to examine the cause of death over the following 11–18 years and compared primary to secondary cigar smokers (those who had switched to cigars from cigarettes). The type of cigar was not reported. They also measured carboxyhemoglobin levels and found that secondary cigar smokers had higher levels than primary smokers (1.0 percent versus 0.9 percent), with both being much lower than that of cigarette smokers (4.6 percent). These data indicate that smoke inhalation on average was much lower in cigar than cigarette smokers. However, some cigar smokers reported moderate to deep inhalation and had carboxyhemoglobin levels similar to those in cigarette smokers. Ischemic heart disease mortality data were presented only for combined cigar and pipe smokers. Primary cigar/pipe users had no higher mortality than lifelong nonsmokers, while secondary cigar/pipe users had a slight but not significant increase in mortality risk (RR 1.29; 95 percent CI: 0.88–1.99). The relative mortality of current cigarette smokers compared to never-smokers was 2.27 (1.81–2.84).

Iribarren et al. (1999) conducted a prospective cohort study in the Kaiser Health system in California of 17,774 men 30–85, followed for 25 years. The group included 1,546 men who currently smoked cigars but never smoked cigarettes. No data were available on type of cigar, but it

was estimated that 21 percent smoked large cigars.⁵ CVD, both nonfatal and fatal, was determined from hospital discharge diagnoses. In a multivariate analysis, cigar smoking was associated with a significantly increased risk of coronary heart disease compared to never-smokers (RR: 1.27; 1.12–1.45). Risk estimates for cigar smoking and ischemic stroke, hemorrhagic stroke, and peripheral arterial disease were nonsignificant, although the case numbers were relatively small. Compared to never-smokers, cigar smokers who used fewer than 5 per day had a lower RR for coronary heart disease (RR: 1.20; 1.03–1.40) compared to those who smoked more than 5 (RR: 1.56; 1.21–2.01). The study could not distinguish occasional versus daily cigar smokers.

CPS-II was another prospective cohort study of 121,278 male primary cigar smokers over age 30 who smoked at least one cigar per day (Jacobs et al., 1999). They were followed between 1982 and 1991, during which time 2,508 deaths occurred. In a multivariable analysis, the IRR for coronary heart disease mortality was 1.30 (1.05–1.62) for current cigar smokers aged 30–74 but was not significantly increased for those 75 or older. Analysis by amount smoked found a significant increase in mortality for those smoking two or more cigars daily, but no increase among those smoking one cigar per day. Analysis by duration found an increase in mortality for those who had smoked for 25 or more years but not for less than 25 years. The mortality risk was also higher among those who reported inhaling compared to those who did not.

Christensen et al. (2018) identified the causes of death based on ICD-10 codes, including circulatory, cardiovascular, and cerebrovascular causes. The age-adjusted risk of circulatory death was significantly increased in former (HR: 1.50; 1.23–1.82) and current (HR: 1.42; 1.12–1.81) cigar smokers compared to never-tobacco users, but these effects became nonsignificant in multivariable analysis controlling for sex, race and ethnicity, education, and survey year. The age-adjusted risk of cardiovascular death was significantly increased in former (HR: 1.56; 1.25–1.94) but not current (HR: 1.24; 0.94–1.62) cigar smokers, with no significant risk after multivariable analysis. No significant increased risk was found for cerebrovascular disease, but the number of deaths was small and considered to be too low to make a robust analysis of daily versus nondaily use (Christensen et al., 2018).

Rostron et al. (2019) studied morbidity associated with current primary cigar smokers age 35 or more using NHIS data between 2000 and 2015. Current use was defined as every day or some days. Health condi-

___________________

⁵ The study authors reported that, based on a previous study using this dataset in 1998 in a smaller group of men (examined between 1979 and 1985), 70 percent had smoked for at least 10 years, and 21 percent smoked large cigars (see Iribarren et al., 1998).

tions were based on self-report or ever having a heart condition (angina, coronary heart disease, heart attack, other heart disease, or stroke). Current primary cigar smoking was not associated with an increased risk of heart attack or stroke. However, former primary cigar smoking was associated with an increased risk of heart conditions (adjusted prevalence ratio [aPR]: 1.33; 1.03–1.72) and stroke (aPR: 2.42; 1.57–3.75) compared to never-smokers. The authors speculate that former cigar use might be explained by smoking cessation in response to disease onset.

Inoue-Choi et al. (2019) used NHIS data from 1991 to 2010 to examine tobacco-related mortality. The analysis included 1,592 exclusive cigar users, but cigar type was not determined. The risk of coronary heart disease mortality was more strongly associated for daily (HR: 1.32; 0.69–2.50) than nondaily (HR: 1.21; 0.57–2.56) cigar smokers compared to never-smokers. For cerebrovascular disease death, neither daily nor nondaily cigar use was associated with increased risk compared to never-tobacco users. A limitation of this study was the relatively small number of deaths (Corrigendum, 2019; Inoue-Choi et al., 2019).

CVD Summary

Cigar smoke is similar to cigarette smoke and would be expected to produce similar cardiovascular toxicity if the extent of exposure is similar. Smoking a single premium cigar has been shown to produce similar acute cardiovascular effects as smoking a cigarette. The effects of long-term cigar use are expected to depend on depth of inhalation and frequency of product use. The studies measuring CO levels suggest that cigar smokers on average inhale much less smoke than cigarette smokers and that secondary cigar smokers inhale more than primary cigar smokers.

Unfortunately, epidemiologic studies to date generally do not report the type of cigar smoked, and the few that did reported on percent of large cigars smoked but did not differentiate premium cigars from other large cigars. Overall, primary cigar smoking appears to be associated with a small but significant increased risk of cardiovascular morbidity and mortality. Cardiovascular risk is generally higher in secondary cigar smokers but still much less than continuing cigarette smokers. There is evidence that the number of cigars smoked per day and depth of inhalation are related to cardiovascular risk. Data are limited on the risks of nondaily cigar smoking, and the available data suggest that the associated cardiovascular risk is low.

Based on what is known about premium cigar product characteristics and the biological mechanism of CVD risk, that risk is likely to be significant if a person smokes premium cigars daily, although generally less than that of cigarette smoking. If a person inhales premium cigar smoke,

the risk would be greater. Exposure to secondhand cigarette smoke by nonsmokers increases the risk of coronary heart disease in general and also of acute coronary events and stroke (Vanker et al., 2017). While it was not studied explicitly, based on such evidence, one would predict that even occasional cigar use would present a similar risk, particularly in individuals with underlying CVD.

A major research need is assessment of the type of cigar, frequency of use, and inhalation when studying disease risk. In addition, cardiovascular risk needs to be assessed in people with existing vascular disease, as this population would be especially vulnerable to adverse effects of acute short-term smoke or nicotine exposure.

LUNG CANCER AND RESPIRATORY DISEASE

Lung Cancer

Cigarette smoking is the main lung cancer risk factor, increasing the risk of incidence and mortality for all major lung cancer histological types. Current and former smokers have considerably higher risks than never-smokers. Lung cancer risk increases as a function of cigarette smoking duration, intensity (usually measured as cigarettes per day), and cumulative exposure (usually measured as pack-years) and decreases as a function of years since quitting (Hecht, 2012; HHS, 2014; Rachet et al., 2004; Remen et al., 2018; Tammemägi et al., 2013).

Cigar use has also been shown to be associated with lung cancer risk, with higher intensity and longer durations associated with higher risk. Most epidemiological studies evaluating the relationship between cigar use and lung cancer have focused on overall cigar use, with no distinction by type. Some studies have evaluated the risk of exclusive cigar use, with others considering cigar and cigarette smoking or cigar and pipe use combined. The literature search found no studies evaluating the specific lung cancer risks from premium cigar use.

Studies of the association between cigar use and lung cancer through 1997 were previously reviewed (NCI, 1998). The evidence was sufficient to conclude that a causal relationship exists between regular cigar use and lung cancer but that the risk was lower than for regular cigarette smokers. It also concluded that lung cancer mortality risk increased as a function of the number of cigars smoked per day and with the depth of inhalation. No consideration was given to variations of risk by cigar type or between different lung cancer histologies.

Recent studies not covered in the 1998 review are summarized here. Boffetta et al. (1999) conducted a case-control study of the association between cigar, cigarillo, pipe, and cigarette smoking with lung cancer

incidence, overall and by histology, in users from seven European areas. The study found lung cancer odds ratio (OR) for exclusive cigar and cigarillo use of 9.0 (95 percent CI: 5.8–14.1) versus nonsmokers of any tobacco product, comparable with the OR of exclusive cigarette smoking of 14.9 (95 percent CI: 12.3–18.1). The study found a dose–response relationship for duration and cumulative exposure for cigar and cigarillos, with similar dose–response relationships for all four products considered. In particular, the analysis found a 3.3 (95 percent CI: 1.8–6.0) lung cancer OR per one log-unit increase in cumulative exposure (measured as grams per day per year, with one cigar assumed to have 4 grams of tobacco on average) of cigars and cigarillos. For cigarette smoking, the odds of lung cancer were also estimated to increase by 3.3 (95 percent CI: 3.1–3.6) per one log-unit increase in cigarette pack-years. An effect was also reported for inhalation of cigar and cigarillo smoke. The authors concluded that the lower overall risk of lung cancer among smokers of cigars and cigarillos compared with cigarette smokers might be due to the lower levels of consumption (intensity) of cigar users. With regards to lung cancer histological type, the risk of squamous cell carcinoma increased along with cumulative consumption of either cigars or cigarillos. Among heavy smokers, the risk of small cell carcinoma was higher than for adenocarcinoma.

Shapiro et al. (2000) conducted a longitudinal study of the relationship between cigar use (measured at baseline) and cancer mortality in CPS-II that did not differentiate cigars by type and excluded those who ever smoked cigarettes or pipes. Current cigar smoking at baseline was associated with increased risk of lung cancer death, with an IRR of 5.1 (95 percent CI: 4.0–6.6). Former cigar smokers were also found to have a higher risk of lung cancer with an IRR of 1.6 (95 percent CI: 1.2–2.4) versus never-users. Strong dose–response relationships by intensity (cigars per day) and duration (25 years or more) were found. Lung cancer mortality risks were considerably higher for cigar users reporting inhalation (IRR: 11.3; 95 percent CI: 7.9–16.1 versus never-users), than for those not reporting inhalation (IRR: 3.3; 95 percent CI: 2.3–4.7 versus never-users). Current cigar smokers at baseline reporting 1–2 cigars per day did not have statistically significant higher risks than never cigar smokers (IRR: 1.3; 95 percent CI: 0.7–2.4).

McCormack et al. (2010) evaluated the association of cigar and pipe smoking with lung cancer incidence in the prospective European Prospective Investigation into Cancer and Nutrition (EPIC) cohort (pooled data from eight cohort centers). Exclusive current cigar smokers were found to have a lung cancer incidence HR of 3.9 (p < 0.05) relative to never-smokers. For comparison, exclusive cigarette smokers were found to have an HR of 32 (p < 0.05). Ever exclusive cigar use was not significantly associated with lung cancer risk (HR: 2.4; 95 percent CI: 0.7–8.2),

but ever exclusive cigarette smoking was significantly associated (HR: 15.2; 95 percent CI: 10.0–23.4). Lung, upper aerodigestive tract (UADT), and bladder cancer (BC) combined risk among exclusive cigar smokers increased by depth of inhalation, duration of use, and intensity (cigars per week). Risk among former cigar users increased by the age at smoking cessation. Notably, exclusive cigar users reporting inhalation were found to have considerably higher lung, UADT, and BC risk (HR: 7.5; 95 percent CI: 3.0–18.8 versus never-smokers), than those reporting no inhalation (HR: 1.8; 95 percent CI: 0.7–4.6). The authors concluded that the lower cancer risk of cigar and pipe smokers as compared to cigarette smokers is explained by lesser degree of inhalation and lower smoking intensity (McCormack et al., 2010).

Lee et al. (2012) conducted a meta-analysis of the relationships among cigarettes, pipes, and cigars and lung cancer and found a relationship with lung cancer risk for ever exclusive cigar smoking (random-effects RR: 2.92; 95 percent CI: 2.38–3.57), current exclusive cigar smoking (RR: 4.67; 95 percent CI: 3.49–6.25), and former exclusive cigar smoking (RR: 2.85; 95 percent CI: 1.45–5.61). For comparison, the analysis found a lung cancer relationship for ever exclusive cigarette smoking (RR: 6.36; 95 percent CI: 5.33–7.59), current exclusive cigarette smoking (9.57; 95 percent CI: 7.90–11.59), and former exclusive cigarette smoking (4.22; 95 percent CI: 3.29–5.40).

Malhotra et al. (2017) evaluated the association between cigar and/or pipe smoking and cancer incidence risk in men in a pooled analysis of five prospective cohorts from the NCI Cohort Consortium in Australia, Netherlands, and the United States (N = 524,400). The study found that ever cigar and/or pipe users were at significantly increased risk for lung cancer incidence versus never-smokers of cigarettes, cigars, and pipes. In particular, exclusive ever cigar users were found to be at higher risk of lung cancer (HR: 2.73; 95 percent CI: 2.06–3.70).

Christensen et al. (2018) evaluated the association of cigarette, cigar, and pipe use with cause-specific mortality, including lung cancer, in the NLMS (N = 357,420). Exclusive current cigar users were at high risk of lung cancer mortality (HR: 3.26; 95 percent CI: 1.86–5.71). Daily exclusive cigar users had a statistically significant higher risk of lung cancer mortality (HR: 4.18; 95 percent CI: 2.34–7.46), but nondaily exclusive cigar users did not (HR: 0.74; 95 percent CI: 0.08–7.26). In a similar study, Inoue-Choi et al. used data from the linked mortality follow-up of the NHIS to evaluate the association among exclusive cigarette, smokeless tobacco, pipe, and cigar use with overall and cause-specific mortality, including lung cancer mortality (Corrigendum, 2019; Inoue-Choi et al., 2019). For exclusive cigar users, the HR for lung cancer mortality was elevated but not statistically significant (HR: 1.87; 95 percent CI: 0.53–6.55). However,

the authors cautioned about the interpretation of these results given the relatively small sample size of cigar users and the small number of cancer deaths observed in this group.

Lung Cancer Summary

As noted throughout this report, cigar smoke is similar to cigarette smoke and would be expected to produce similar lung cancer risks if the magnitude of exposure is similar. Like cigarette smoking, cigar smoking has been shown in multiple epidemiological studies to result in considerable lung cancer incidence and mortality risk and to have strong dose–response relationships with intensity, duration, and cumulative exposure. Risk for former cigar smokers is highest for those quitting use at older ages. The lung cancer risk from cigar use is considerably higher for users who report inhalation.

Limited information is available regarding the risks by specific cigar types, with only one study reporting risks for cigars versus cigarillos. The literature search found no studies reporting specific risks for premium cigar users. However, based on the cigar literature, it is expected that daily or frequent long-term use of premium cigars would result in considerable lung cancer risk.

A major research need is the consideration of type of cigar, including premium cigars, as well as the frequency of use, duration, intensity, cumulative exposure, and pattern of inhalation when studying the associations with lung cancer. In addition, the existing literature does not estimate the associations between cigar use and specific lung cancer histological types.

Other Respiratory Diseases

Despite several studies examining the relationship of cigar use with lung cancer, only a few have studied their impact on other respiratory diseases. The literature on the relationship between cigar use and chronic obstructive pulmonary disease (COPD) was reviewed through 1997 (NCI, 1998). Based on two studies, the authors concluded that while the association between cigar and COPD risk is less striking than that for cigarette smoking, the studies reviewed support the conclusion that cigar smoking can cause COPD in users who inhale deeply. The 1998 review also concluded that the reduced inhalation of tobacco smoke by cigar users probably explains the lower risks of COPD and lung cancer among cigar smokers compared to cigarette smokers (NCI, 1998).

The committee’s literature search found only four studies published after 1997 reporting on the association between cigar use and COPD. A few additional studies evaluated the association of cigar use with asthma

and respiratory symptoms, such as wheezing. None of these studies evaluated risks specific for premium cigar users.

Jimenez Ruiz et al. (2002) evaluated the prevalence of COPD in different tobacco use groups in Spain using a cross-sectional, nationally representative sample. Data from 4,035 individuals aged 40–69 were included, dividing the sample into noncurrent smokers, current exclusive cigar smokers, current exclusive cigarette smokers, dual smokers, current exclusive cigar smokers who used to smoke cigarettes, and current cigarette smokers who used to smoke cigars. The analysis found that all current use groups had statistically significantly higher prevalence of COPD (diagnosed through spirometry) and respiratory symptoms (cough and expectoration) versus noncurrent smokers. However, the prevalence of COPD and respiratory symptoms was higher in those reporting either current cigarette smoking, exclusively or dual, or in cigar smokers who used to smoke cigarettes, than in those reporting exclusive cigar use.

Mannino et al. (2000) examined the prevalence of COPD or asthma (obstructive lung disease), low lung function, and respiratory symptoms by tobacco use group in the National Health and Nutrition Examination Surveys (NHANES) from 1988 to 1994. Respondents were classified as never-smokers, current cigarette smokers, former cigarette smokers, or pipe or cigar smokers. Low lung function was defined as a forced expiratory volume in 1-second forced vital capacity (FEV1/FVC) of less than 0.7 and an FEV1 less than 80 percent of the predicted value. The study found an age-adjusted prevalence of obstructive lung disease of 12.5 percent among current cigarette smokers, 9.4 percent among former cigarette smokers, 3.1 percent among pipe or cigar smokers, and 5.8 percent among never-smokers.

Christensen et al. (2018) evaluated the association of cigarette, cigar, and pipe use with cause-specific mortality, including COPD, in NLMS (N = 357,420) and found a borderline nonsignificant association between current exclusive cigar use overall and COPD (HR: 2.44; 95 percent CI: 0.98–6.05) but a significant association between current exclusive daily cigar use and COPD (HR: 3.29; 95 percent CI: 1.33–8.17).

Rodriguez et al. (2010) studied the association of pipe and cigar use with cotinine levels, lung function, and airflow obstruction in the MultiEthnic Study of Atherosclerosis. Participants reporting a history of pipe, cigar, or cigarette use were classified as exclusive ever users of pipes or cigars combined, exclusive ever users of cigarettes, or ever users of both products. Lung function, measured as FEV1 or FEV1/FVC, decreased among participants with a history of pipe or cigar smoking only, cigarette smoking only, and pipe or cigar and cigarette smoking, compared to never-smokers. However, the decrement was modest and not statistically significant among the 55 participants who smoked pipes or cigars only.

The odds of airflow obstruction increased in all tobacco use groups compared with never-smokers: pipes or cigars only (OR: 2.31; 95 percent CI: 1.04–5.11), cigarettes only (OR: 2.01; 95 percent CI: 1.31–3.08), and pipes or cigars and cigarettes (OR: 3.43; 95 percent CI: 1.75–6.71). Greater cigar-years (product of number of years smoked or duration times cigars per day) were associated with a decrement in lung function, which was statistically significant for FEV1/FVC ratio; -0.2 (-0.3, -0.05) decrease in FEV1/FVC ratio per 10 cigar-years (duration of use times number of cigars per day). The authors concluded that pipe and cigar smoking measurably increase the risk of COPD.

Three studies evaluated associations between cigar and other tobacco product use with asthma. Jones et al. (2006) compared the prevalence of tobacco use among high school students with and without self-reported asthma in the 2003 Youth Risk Behavior Survey and found that those with current asthma used cigarettes at higher rates than those without asthma, but the rate of cigar use was similar between those with and without current asthma (OR: 1.0; 95 percent CI: 0.9–1.2). Among students with current asthma, those who had an asthma episode or attack were significantly more likely than those who had not to report lifetime daily cigarette use (OR: 1.5; 95 percent CI: 1.1–2.1), current frequent cigarette use (OR: 1.6; 95 percent CI: 1.04–2.6), and current cigar use (OR: 1.6; 95 percent CI: 1.03–2.6). Lappas et al. (2016) compared the immediate effects of cigar smoking on respiratory mechanics and exhaled biomarkers between young smokers with and without mild asthma. Participants with mild asthma were recruited from an outpatient lung function clinic. The results suggest that cigar smoking has immediate effects on pulmonary function, affecting exhaled CO, multifrequency respiratory system impedance, and other outcomes, with mild asthma being associated with a higher increase of peripheral airway resistance (frequency dependence of resistance) after cigar smoking. Veldhuis et al. (2021) evaluated the association of self-reported asthma, sexual identity, and inhaled substance use, including cigar use, among U.S. adolescents and found that cigarettes, cigars, and electronic vapor products were all associated with asthma in both female (cigar relative risk ratio [RRR] 1.58; 95 percent CI: 1.28–1.96) and male (cigar RRR of 1.35; 95 percent CI: 1.14–1.62) adolescents. Similar risks were estimated for all tobacco products.

Schneller et al. (2020) evaluated the association between different tobacco product use and self-reported wheezing symptoms among U.S. adults from the Population Assessment of Tobacco and Health (PATH) study. Significant higher odds of ever had wheezing or whistling in the chest at any time in the past were observed among current cigarette (adjusted odds ratio [aOR]: 2.62; 95 percent CI: 2.35–2.91), electronic nicotine delivery systems (ENDS) (1.49; 95 percent CI: 1.14–1.95), and polytobacco (2.67; 95 percent CI: 2.26–3.16) users compared with noncurrent users. However, no significant association was found for cigar use.

extends from the base of the gingival sulcus to the gingival margin, and the interdental gingiva fills the space between the teeth (Fiorellini, 2019).

A healthy gingival margin is positioned approximately 1.5–2.0 mm coronal to the cemento-enamel junction (where the enamel on the tooth crown meets the root) (Lindhe, 2015); the sulcus probing depth is ≤3 mm, and does not bleed when probed (Do, 2019). The sulcus base is formed by junctional epithelium, which joins the gingival connective tissue to the tooth surface. Healthy gingiva should be pink, well adapted to the teeth, stippled on the surface, and tightly bound to the alveolar bone and tooth roots (Do, 2019).

Periodontitis

“Periodontitis is a chronic multifactorial inflammatory disease associated with dysbiotic plaque biofilms and characterized by progressive destruction of the tooth-supporting apparatus” (Papapanou et al., 2018, p. 1). Primary features of peritonitis include presence of periodontal pocketing, gingival bleeding, and the loss of periodontal tissue support (manifested through clinical attachment loss and radiographically assessed alveolar bone loss) (Papapanou et al., 2018). Periodontitis is considered a major public health problem due to its high prevalence “and because it may lead to tooth loss and disability, negatively affect chewing function and aesthetics, be a source of social inequality, and impair quality of life” (Papapanou et al., 2018, p. 1). In addition, periodontal inflammation is associated with several chronic conditions, including CVD, diabetes and its management, chronic kidney disease, rheumatoid arthritis, and pregnancy complications (Bui et al., 2019; Hajishengallis and Chavakis, 2021; Kapellas et al., 2019; Liccardo et al., 2019; Mankia et al., 2019; Martinez-Herrera et al., 2017; Moliner-Sánchez et al., 2020; Rodríguez-Lozano et al., 2019; Sanz et al., 2018).

Mechanisms of Tobacco-Smoke-Caused Periodontal Disease

No studies are specific to the biological mechanisms of periodontitis associated with cigar use. Multiple lines of investigation exist on the biologic mechanisms involved in periodontitis due to cigarette smoking, although smoking-related periodontal pathogenesis is not yet fully understood. To the extent that smoke from combusted premium cigars contains similar agents to mainstream cigarette smoke, the mechanisms associated with cigar-associated periodontitis are likely to be similar to those involved in cigarette-related periodontitis.

Oral microorganisms have been established as major factors in the pathogenesis of periodontitis for more than half a century. The oral cavity

is a complex ecosystem that can harbor hundreds of bacterial species as well as other microbes that normally act as symbiotic communities with the host (Lasserre et al., 2018). Periodontal disease and health are more likely to be associated with qualitative or quantitative shifts in microbiome within periodontal biofilms rather than the presence or absence of specific pathogenic bacteria. It is hypothesized that soft tissue and alveolar bone destruction involves both toxins and proteases produced by the bacteria and hyperresponsiveness and reactivity of immune system components, including the production of cytokines and prostaglandins (HHS, 2004). Multiple studies have found that cigarette smoking affects the composition of the oral microflora (Apatzidou et al., 2005; Hanioka et al., 2000; Jiang et al., 2020; Kubota et al., 2011; Moon et al., 2015; van Winkelhoff et al., 2001). It also affects humoral and cell-mediated immune responses, which may increase susceptibility to periodontitis (Loos et al., 2004; Palmer et al., 2005; Ryder, 2007), and appears to alter the periodontal inflammatory response (Dietrich et al., 2004).

Based on findings from animal studies, it was hypothesized that the peripheral vasoconstrictive effect of nicotine reduces gingival blood flow and thereby impairs the delivery of oxygen and nutrients to the tissue (Clarke and Shephard, 1984). However, subsequent evidence from human studies does not support that hypothesis (Silva, 2021). A comprehensive review by Silva (2021) does suggest that chronic tobacco exposure causes long-term microvascular dysfunction, which may play a role in the progression of periodontitis. It has long been observed that smokers tend to exhibit less gingival bleeding than nonsmokers, even when controlling for bacterial plaque levels (Bergström and Boström, 2001; Dietrich et al., 2004; Rivera-Hidalgo, 2003). However, this decrease may be more related to suppression of an inflammatory response than to reduced gingival blood flow (Silva, 2021).

Nicotine can be stored in and released from periodontal fibroblasts and may affect their morphology and ability to attach to root surfaces (Hanes et al., 1991; James et al., 1999; Raulin et al., 1988; Tanur et al., 2000). Thus, it is possible that smoking impairs the ability of periodontal tissues to repair damaged junctional epithelium (HHS, 2004). Substantial evidence indicates that smoking impairs wound healing and compromises outcomes following surgical or nonsurgical periodontal therapy (Ah et al., 1994; Boström et al., 1998; Grossi et al., 1996, 1997; Kaldahl et al., 1996; Kinane and Radvar, 1997; Machtei et al., 1998; Newman et al., 1994; Palmer et al., 1999; Papantonopoulos, 1999; Preber and Bergström, 1990; Preber et al., 1995; Renvert et al., 1998; Rosenberg and Cutler, 1994; Söder et al., 1999; Tonetti et al., 1995; Trombelli and Scabbia, 1997). Although the exact mechanisms are not yet known, the various factors produce increased tissue destruction and diminished healing response, with a net effect of periodontal tissue breakdown.

(4.0 versus 1.9; p = .0006). The analyses for cigar/pipe smoking did not exclude or control for cigarette smoking.

Vora and Chaffee (2019) analyzed cross-sectional data on adults from the first wave (2013–2014) of PATH. The outcomes in that study were based on participants’ self-report in response to two questions: “Have you ever been told by a dentist, dental hygienist, or other health professional that you have gum disease?” and “Have you ever had treatment for [gum disease, your gums] such as scaling and root planing, sometimes called deep cleaning.” Cigar types included traditional cigars, cigarillos, and filtered cigars (results combined). In multivariable modeling that included age, sex, race/ethnicity, educational attainment, employment, use of dental services, and history of diabetes, current exclusive cigar smokers were more likely than adults who never used tobacco to report a diagnosis of gingival disease (OR: 1.9; 95 percent CI: 1.4–2.7) or treatment for it (OR: 1.5; 95 percent CI: 1.2–2.0). The strength of association between current cigar use and self-reported gum disease diagnosis or treatment was similar to those for current cigarette use (ORs: 2.2 and 1.5, respectively).

These three epidemiologic studies on cigar smoking consistently found more prevalent or incident periodontitis among cigar users than among nonsmokers. However, none of those studies were specific to premium cigars. One (Albandar et al., 2000) combined current and former use of cigars or pipes and did not exclude cigarette smoking in its measure of exposure, and one (Vora and Chaffee, 2019) used an outcome measure based on self-report, which may have low sensitivity compared with clinically determined disease status (Blicher et al., 2005; Gilbert and Nuttall, 1999; Yamamoto et al., 2009).

Periodontal Diseases Summary

Very few human studies have estimated the risk of periodontitis associated with cigar smoking, and none explicitly studied premium cigars. However, three epidemiologic studies consistently found elevated odds of periodontitis compared with nonsmokers. Those findings are consistent with the relatively large body of literature on cigarette smoking and periodontitis, which is sufficient to reach a strong conclusion that cigarette smoking is a cause. Similarly, no known mechanistic studies exist specific to cigars in general or premium cigars in particular. However, the evidence is substantial that cigarette smoking is associated with changes in the oral microbial profile, causes disruption to humoral and cell-mediated immune function, degrades bone and soft tissue integrity, and impairs tissue repair. To the extent that combustion of premium cigars produces many of the same toxic agents as in mainstream cigarette

smoke, the various biologic mechanisms are likely involved in cigar-associated periodontitis.

Cancers of the Oral Cavity, Head, and Neck

Mechanism

The major established pathways of cancer causation by cigarette smoking involves the exposure to carcinogens, the formation of covalent bonds between the carcinogens and DNA (DNA adduct formation), and the resulting accumulation of permanent somatic mutations in critical genes, which lead to clonal outgrowth and, through accumulation of additional mutations, to development of cancer (HHS, 2010).

Human Studies

Seven case-control studies that investigated the association between cancers of the oral cavity and pharynx and cigar use have been published since the late 1980s (Blot et al., 1988; Franceschi et al., 1990, 1992; Garrote et al., 2001; Merletti et al., 1989; Schlecht et al., 1999; Spitz et al., 1988). Five were included in a narrative review (NCI, 1998). Two were conducted in the United States, three in Italy, one in Cuba, and one in Brazil. None of the seven specifically reported data on premium cigars; four (Blot et al., 1988; Franceschi et al., 1990, 1992; Garrote et al., 2001) combined use of cigars or pipes in their analyses; and two also included cancers of the larynx among the cancer outcomes in the main analysis (Schlecht et al., 1999; Spitz et al., 1988). Most studies explicitly controlled for cigarette smoking either through exclusion of concurrent cigarette smokers or by adjustment in multivariable modeling, but control for smoking was not clear in one of the studies (Spitz et al., 1988). All seven studies found a significant positive association between cigar use and cancers of the oral cavity or pharynx, with aOR estimates of 1.9–21.9. One study reported a dose-dependent association among current cigar smokers (Garrote et al., 2001), and another found that the odds of UADT cancers declined with the number of years since quitting cigar use (Schlecht et al., 1999).

A recent systematic review and meta-analysis of 13,935 cases and 18,691 controls from 13 case-control studies conducted in multiple regions of the world examined the association between cigar use (type not specified) and cancers of the head and neck, including of the oral cavity, oropharynx, hypopharynx, larynx, and other nonspecified sites (Wyss et al., 2013). Among persons who had never smoked cigarettes, those who had ever used cigars were at elevated risk compared with those who never

used cigars (OR: 2.54; 95 percent CI: 1.93–3.34). Among cigar smokers who never smoked cigarettes, the odds of head and neck cancer significantly increased with the number of cigars per day, duration of cigar use, and cumulative cigar-years (p for trend <.0001 for all three). In site-specific analysis among persons who never smoked cigarettes, cigar use was associated with increased odds of cancer of the oropharynx (OR: 2.31; 95 percent CI: 1.54–3.45) and all other cancer sites.

Two large U.S. prospective studies estimated the risk for death due to cancers of the oral cavity or pharynx associated with cigar use at baseline. Shanks and Burns reported findings from CPS-I (NCI, 1998). CPS-I collected baseline data in 1959 and tracked cause-specific mortality for up to 13 years; it conducted cigar analyses for white men and classified users as either primary cigar smokers (those who never used cigarettes or pipes) or secondary cigar smokers (those who had formerly used cigarettes or pipes but only used cigars at the time of the baseline data collection). Overall, white men who were primary cigar smokers at baseline had an age-adjusted IRR for death due to oral or pharyngeal cancer of 7.92 (95 percent CI: 5.12–11.69) relative to never-smokers, and secondary cigar smokers had an age-adjusted IRR of 6.58 (95 percent CI: 2.83–12.97). Analysis of CPS-I data also revealed a dose-dependent risk among primary cigar smokers (IRR: 2.12; 95 percent CI: 0.43–6.18 among men who smoked 1–2 cigars per day to IRR: 15.94; 95 percent CI: 8.71–26.75 for men who smoked 5 or more per day). A similar pattern was reported in secondary cigar smokers (IRR: 4.39–13.73, despite no data for men who smoked 3–4 cigars per day). That study also provided RR estimates by reported depth of inhalation. The risk of mortality due to cancer of the oral cavity or pharynx among primary cigar smokers increased consistently with reported depth of inhalation (IRR: 6.98; 95 percent CI: 4.13–11.03 among men who reportedly did not inhale to IRR 27.88; 95 percent CI: 5.60–81.46 among those who reported moderate to deep inhalation). The pattern was identical among secondary cigar smokers.

CPS-II was a prospective cohort study that enrolled 1.2 million men and women in 1982 and tracked cause-specific mortality for up to 12 years (Shapiro et al., 2000). Similar to CPS-I, analyses of mortality among cigar smokers were limited to men. In CPS-II the mortality IRR for cancers of the oral cavity or pharynx was 4.0 (95 percent CI: 1.5–10.3) among current cigar smokers and 2.4 (95 percent CI: 0.8–7.3) among former cigar smokers. The IRR estimates were higher among men who reported inhalation of cigar smoke (IRR: 6.5; 95 percent CI: 1.4–29.2) than among those who reportedly did not inhale (IRR: 3.2; 95 percent CI: 0.9–11.0). The data were too sparse for detailed analysis by number of cigars per day.

Iribarren et al. (1999) examined UADT cancers in exclusive cigar smokers (cigar type not reported). Among men who had never smoked

cigarettes and did not currently smoke a pipe, those who currently smoked cigars at baseline experienced an age-adjusted incidence rate of UADT cancers of 2.0 per 10,000 person-years. Compared with men who did not smoke cigars, the adjusted RR was 2.02 (95 percent CI: 1.01–4.06). The association was stronger when analysis was limited to cancers of the oropharynx (adjusted RR: 2.61; 95 percent CI: 1.18–5.76).

In a recent meta-analysis, Malhotra et al. (2017) found that cigar (type not reported) and/or pipe smokers were at elevated risk for cancers of the head and neck (HR 1.51; 95 percent CI: 1.22–1.87). In a subgroup analysis that included the two cohort studies with the most detailed data on frequency and duration of tobacco use, exclusive cigar smokers with no history of cigarette smoking had an elevated risk of these cancers (HR 2.59; 95 percent CI: 1.21–5.58).

Pooled data from the EPIC study were used to investigate UADT cancer incidence rates among men who smoked cigars (type of cigar not reported) (McCormack et al., 2010). Men who ever exclusively smoked cigars had an elevated HR relative to those who never smoked (HR: 4.0; 95 percent CI: 1.7–9.4). Men who had quit cigarette smoking and became current cigar smokers had an HR for UADT cancer (HR: 8.2; 95 percent CI: 4.1–16.7) that was comparable to that observed for exclusive cigarette smokers (HR: 8.9; 95 percent CI: 3.1–6.6).

Cancers of the Oral Cavity, Head, and Neck Summary

Consistent data from all identified cohort and case-control studies indicate a significantly elevated risk for oral and pharyngeal cancer associated with cigar use, with evidence of a dose-dependent relation. Coupled with biologic mechanisms that likely are very similar to those involved in cigarette-related carcinogenesis, the available evidence strongly supports the conclusion that cigar use is a cause of cancer of the oral cavity and pharynx. Although none of the available studies specifically examined the risk associated with premium cigars, it is very likely that their use also increases the risk for oral and pharyngeal cancer. The level of increased risk will likely depend on the frequency of premium cigar smoking, which is generally lower than that of smoking other types of cigars.

OTHER CANCERS

Cigar use is associated with the risk of other cancers. In particular, the 1998 NCI monograph reviewed the literature on the associations between cigar use and bladder and pancreatic cancer based on the evidence through 1997 (NCI, 1998). It concluded that although a few studies suggested an association between BC risk and cigar use, several other

studies had not found evidence of such association. In contrast, the 1998 review concluded that cigar users have higher rates of pancreatic cancer with increasing risk with higher number of cigars per day, level of inhalation, and age. It also concluded that regular cigar use causes cancers of the lung, oral cavity, larynx, esophagus, and probably pancreas.

Recent studies evaluating the associations of cigar use with bladder, pancreatic, and other cancers since 1997 are discussed next (note that esophageal, bladder, and pancreatic cancer are also discussed in greater detail in the following sections).⁶Andreotti et al. (2017) evaluated the associations between exclusive cigar, cigarillos, and smokeless tobacco ever use with the incidence of several cancers in the prospective Agricultural Health Study (n = 84,015). They found that ever exclusive cigar use at baseline was associated with all cancers (HR: 1.51; 95 percent CI: 1.20–1.90) and smoking-related cancers⁷ (HR: 1.87; 95 percent CI: 1.24–2.82), but no statistical association was found with gastrointestinal cancers (HR: 1.58; 95 percent CI: 0.84–2.98). The study also found a positive association with urinary cancer (i.e., bladder, kidney, and ureter cancers combined) (HR: 2.50; 95 percent CI: 1.27–4.93). Only 76 total cancers in exclusive ever cigar users were available for the study, precluding site-specific analyses. The study found that dual cigarette and cigarillo use is associated with higher risks of overall, smoking-related, and lung cancers than exclusive cigarette smoking but that dual cigar and cigarette smoking has similar risks as exclusive cigarette smoking.

Engeland et al. (1996) evaluated the associations of smoking habits, including cigar smoking, and the incidence of cancers other than lung among 26,000 Norwegian men and women recruited in 1965 and followed through 1993. The cancers studied were urinary, bladder, kidney, pancreas, upper digestive and respiratory tract (i.e., head and neck and esophageal cancer combined), uterine, cervix, stomach, colon, rectum, breast, corpus uteri, ovary, and prostate, and leukemia. No association was found between cigar use and any of the studied cancers.

Malhotra et al. (2017) evaluated the association between cigar and/or pipe smoking and cancer incidence risk in men and found that ever cigar and/or pipe users were at significantly increased risk for head and neck cancer, lung cancer, and liver cancer versus never-smokers of cigarettes, cigars, and pipes. The risk of smoking-related cancers combined and of all cancers combined was also found to be significantly higher in ever cigar

___________________

⁶ Two studies were not included in the review: Efird et al. (2004) (methods say ever cigar, but table 2 suggests current cigar; excluded due to inconsistency); Sasco et al. (2004) (just one paragraph on other tobacco products).

⁷ The smoking-related cancers in the study included bladder, colon, cervix, esophagus, kidney, larynx, lip, liver, lung, myeloid leukemia, nasal and sinus, oral cavity, pancreas, pharynx, rectum, stomach, tongue, ureter, and uterus.

and/or pipe smokers. Exclusive ever cigar smokers were found to be at higher risk of head and neck cancer (HR: 1.40; 95 percent CI: 0.98–2.00), lung cancer (HR: 2.73; 95 percent CI: 2.06–3.60), smoking-related cancers (HR: 1.47; 95 percent CI: 1.34–1.61), and all cancers combined (HR: 1.07; 95 percent CI: 1.02–1.16).

McCormack et al. (2010) evaluated the association of cigar and pipe smoking in the prospective EPIC cohort. The cancers evaluated included lung, UADT, bladder, liver, stomach, pancreas, kidney, colorectal; lung, UADT, and bladder combined; and all these tobacco-related cancers combined. Exclusive current cigar smokers were found to have higher risk compared to never-smokers of lung (HR: 3.9; p < 0.05); UADT (HR: 3.5; p < 0.05); lung, UADT, and bladder combined (HR: 2.6; p < 0.05); and all tobacco-related cancers combined (HR: 1.6; p < 0.05). Other cancers were not associated with exclusive current cigar use or had no cases among exclusive current cigar users to measure any association (pancreatic). Ever exclusive cigar use was found to be significantly associated with UADT risk (HR: 4.0; 95 percent CI: 1.7–9.4), lung, UADT, and bladder cancers combined (HR: 2.2; 95 percent CI: 1.3–3.8), and all tobacco-related cancers combined (HR: 1.3; 95 percent CI: 1.0–1.8). Other cancers were not associated with ever exclusive cigar use. Lung, UADT, and bladder cancer combined risk among exclusive cigar smokers increased by depth of inhalation, duration of use, and intensity (cigars per week). Risk among former cigar users increased by the age at smoking cessation (see study description in the lung cancer section for additional results).

As explained previously, Shapiro et al. (2000) conducted a prospective study of the relationship between cigar use (baseline) and cancer mortality in CPS-II. In addition to the previously noted increased risk of death from cancers of the lung and oral cavity/pharynx, the study found that current cigar smoking at baseline was associated with an increased risk of death from cancers of the larynx (IRR: 10.3; 95 percent CI: 2.6–41.0). However, no significant associations were found for overall current cigar smoking and esophagus, pancreas, and bladder mortality cancer risk. As with lung and oral cavity/pharynx, cancer mortality risks were considerably higher for cigar users reporting inhalation versus never-users: larynx (IRR: 39.0; 8.4–180.1), pancreas (IRR: 2.7; 95 percent CI: 1.5–4.8), and bladder (IRR: 3.6; 95 percent CI: 1.3–9.9). No association was found for cigar smokers reporting inhalation and esophagus cancer mortality.

As noted earlier, Christensen et al. (2018) evaluated the association of cigarette, cigar, and pipe use with cause-specific mortality, including tobacco-related cancers (e.g., bladder, esophagus, larynx, lung, oral cavity, and pancreas) in the NLMS (n = 357,420). Exclusive current cigar users had an increased risk of tobacco-related cancer mortality (HR: 1.61; 95 percent CI: 1.11–2.32). Exclusive daily cigar users were found to have

higher risk of tobacco-related cancer mortality (HR: 1.80; 95 percent CI: 1.20–2.69), but not exclusive nondaily cigar smokers (HR: 1.08; 95 percent CI: 0.45–2.61).

Two studies evaluated the relationship between cigar use and non-Hodgkin’s lymphoma (NHL). Bracci et al. (2005) addressed the associations between tobacco use, including past-year cigar use, and NHL in a case-control study of HIV-negative NHL patients and population controls from the San Francisco Bay area (N = 1,593 total patients, N = 2,515 total controls). Among men, the study found no associations between cigar use and overall NHL (OR: 1.3; 95 percent CI: 0.54–3.0) but an association with follicular NHL (OR: 2.8; 95 percent CI: 1.1–7.2). Neither female cases nor controls reported exclusive cigar use. Fernberg et al. (2006) considered the associations between tobacco use, including cigar smoking, and the risk of malignant lymphomas in a cohort of 386,000 Swedish construction workers recruited at clinics from 1971 and 1992 and followed through 2000. Smoking cigarettes, pipes, or cigars was not associated with NHL or Hodgkin’s disease. In particular, the study found that smoking one or more than one cigar per day was not related to a higher risk of NHL (IRR: 0.86; 95 percent CI: 0.58–1.27).

Sorahan et al. (1997) evaluated the association between parental tobacco use, including current cigar use among fathers, and childhood cancers in children from the Oxford Survey of Childhood Cancers study (2,587 cancer cases and 2,587 controls). No associations were found between paternal cigar use and childhood cancer (RR: 0.98; 95 percent CI: 0.69–1.40).

Esophageal Cancer

The association between cigarette smoking and esophageal cancer is well established. The risk among current cigarette smokers may be up to 7.5 times higher than nonsmokers (HHS, 2014). Furthermore, many studies have reported a dose–response relationship and a reduction in mortality after quitting cigarettes. In 1998, NCI concluded that cigar use also caused esophageal cancer and, in fact, had similar mortality rates to cigarette use (NCI, 1998). Since this review, few new studies have explored this association, and no studies specifically examined the effect of premium cigars. Presented here are summaries of the results of NCI (1998) and three more recent studies identified in the literature review.

Based on data from CPS-I, the NCI monograph reported that cigar smokers who have never smoked cigarettes (primary cigar users) have an increased risk of developing and dying of esophageal cancer when compared with nonsmokers (NCI, 1998). This association with mortality was replicated across four other prospective cohort studies and six

case-control studies (with a range in IRR of 2.0–6.7). Three of these studies (one case-control and two prospective) combined cigar and pipe use. Furthermore, the association between cigar use and esophageal cancer remained regardless of inhalation. As noted, unlike cigarette smokers, cigar smokers are less likely to inhale, resulting in different patterns of smoke exposure. The NCI monograph noted that this could contribute to the differences in mortality ratios by cause of death observed between cigar and cigarette users. Esophageal and oral cancers have similar mortality ratios for both, while cigarette users have higher mortality ratios for coronary heart disease, COPD, lung cancer, and laryngeal cancer. The association observed between cigar use and esophageal cancer is further supported by a dose–response relationship and biological evidence of carcinogenic agents in cigars affecting risk of esophageal cancer in both rats and humans. Finally, the NCI monograph noted that despite few data on occasional cigar smokers, the risks of esophageal cancer and other causes of death are likely to be greater than in those with no tobacco exposure and less than in regular users.

In 2000, Shapiro and colleagues analyzed data from CPS-II and found a positive, but not significant, association between cigar smoking at baseline and esophageal cancer (IRR 1.8; 95 percent CI: 0.9–3.7). Shapiro and colleagues also investigated cigars per day, inhalation, and years individuals had smoked. No associations were significant, but there was a greater mortality IRR for those who smoked longer than 25 years compared to those who did not (2.2 versus 0.9). Although no associations were significant, current cigar smokers registered only nine esophageal cancer deaths. Many studies included in reviews on this topic suffered from having a small number of esophageal cancer cases, making it difficult to detect true associations.

In 2015, Chang and colleagues undertook a systematic review to identify prospective cohort studies published before June of 2014. They did not identify new U.S. studies compared to the 1998 monograph other than the Shapiro study discussed above. This has further exposed the lack of research on cigars and esophageal cancer.

The last paper identified in the review was a 2017 meta-analysis of five prospective cohort studies. Cohorts were identified in Australia, Netherlands, and the United States (Malhotra et al., 2017). The investigators examined self-reported ever cigar use and predominant cigar use (most tobacco exposure resulting from cigars rather than cigarettes) with esophageal cancer mortality. They found no significant association for ever cigar use (HR: 1.01; 95 percent CI: 0.56–1.84) nor for exclusive cigar (HR: 1.39; 95 percent CI: 0.35–5.47) or predominant cigar users (HR: 1.45; 95 percent CI: 0.37–5.73). Again, this study had very few cases, with only 12 cases for ever cigar use.

___________________

⁸ The following studies identified in the literature search were excluded and not discussed in the review: Boffetta et al., 2008 (review article that only discussed one study of cigar use—that study was discussed separately); Pramod et al., 2020 (review article that only briefly discussed three studies on cigar use—these studies are discussed separately); Zeegers et al., 2002 (pipe and cigar combined); and Zeegers et al., 2004 (review article that discussed the same articles from NCI monograph and articles that combined pipe and cigar smoking).

data from 442,455 men in CPS-I found that compared to never-smokers, the age-standardized IRRs (95 percent CI) for BC mortality were 1.38 (0.89–2.04) for primary cigar smokers, 1.23 (0.56–2.33) for secondary cigar smokers, 3.17 (2.83–3.54) for cigarette smokers, and 2.48 (1.42–4.03) for dual users of cigars and cigarettes.

Shapiro and colleagues examined the association between cigar smoking and death from tobacco-related cancers, including BC, using CPS-II data (Shapiro et al., 2000). Neither current nor former cigar smoking was associated with death from BC in the overall study population (IRR: 1.0; 95 percent CI: 0.4–2.3 for current cigar-only smokers and 1.3; 95 percent CI: 0.7–2.5 for former cigar-only smokers compared to never-smokers). However, mortality from BC was increased, although not significantly, for current cigar smokers who reported smoking ≥3 cigars/day (IRR: 1.9; 95 percent CI: 0.8–4.4) (Shapiro et al., 2000). Note that the number of deaths from BC in this study was low—94 in never-smokers, 10 in former smokers, and 6 in current smokers.

A pooled analysis of data from six case-control studies (2,279 BC cases and 5,268 controls) from Denmark, France, Germany, and Spain was conducted to assess the association between cigar, pipe, and cigarette smoking and BC risk in European men (Pitard et al., 2001). After adjustment for age, center, and occupational exposure, the OR (95 percent CI) for BC was 2.3 (1.6–3.5) for primary cigar smoking, 1.9 (1.2–3.1) for primary pipe smoking, and 3.5 (2.9–4.2) for primary cigarette smoking compared to never-smokers.

A prospective study of 102,395 in the EPIC cohort examined the effects on cancer incidence of exclusive cigar and pipe smoking, and in combination with cigarettes (McCormack et al., 2010). Compared to never-smokers, the HRs (95 percent CI) for BC were 1.5 (0.6–3.5) for ever exclusive cigars smokers, 1.7 (0.9–3.4) for ever exclusive pipe smokers, and 2.9 (2.3–3.7) for ever exclusive cigarette smokers.

Cumberbatch et al. (2016) conducted a meta-analysis of studies published through August 2013 that examined the impact of tobacco exposure on BC incidence and mortality. The authors reported increased BC incidence among cigar smokers compared to never-smokers (RR: 1.62; 95 percent CI: 1.18–2.22). Relative to other tobacco products, the risk of incident BC was similar for pipe smokers (RR: 1.49; 95 percent CI: 1.18–1.88) but lower than the risk for cigarette smokers (RR: 3.37; 95 percent CI: 3.01–3.78). Cigar smoking also had a nonsignificant higher risk of death from BC, but BC mortality was less extensively reported in the literature.

The aforementioned Chang et al. (2015) systematic review found that mortality ratios for BC was 0.94–1.9 for current cigar smoking.

Al-Zalabani et al. (2016) conducted a systematic review and meta-analysis of articles published between 1995 and 2015 that examined modi-

fiable risk factors of primary BC, including cigar smoking. Using data from six studies, the authors reported increased risk of BC incidence among cigar smokers compared to never-smokers (RR: 2.3; 95 percent CI: 1.6–3.5). Relative to other tobacco products, the risk of incident BC was higher for primary cigar smokers than for primary pipe smokers (RR: 1.90; 95 percent CI: 1.2–3.1) and former cigarette smokers (RR: 1.83; 95 percent CI: 1.52–2.14) but lower than the corresponding risk for current cigarette smokers (RR: 3.14; 95 percent CI: 2.53–3.75).

In a pooled analysis of data from five prospective cohorts in the NCI Cohort Consortium, Malhotra et al. (2017) examined the association between exclusive cigar and/or pipe smoking with BC incidence among men. Compared to never-smokers, HRs were 1.14 (95 percent CI: 0.88–1.48) for ever cigar smokers and 1.40 (1.07–1.84) for ever pipe smokers.

In the prospective study by Inoue-Choi et al. (2019)⁹ described above, the authors found that compared to never-tobacco users, HRs for BC mortality were 5.68 (95 percent CI: 0.74–43.69) for current exclusive cigar smokers, 4.65 (95 percent CI: 2.65–8.17) for current exclusive cigarette smokers, and 6.90 (95 percent CI: 1.06–45.14) for current exclusive smokeless tobacco users.

Impact of Intensity and Duration of Cigar Smoking on Bladder Cancer

A paucity of studies have examined the impact of intensity or duration of cigar smoking on risk of BC. Among studies that examined the impact of duration (Boffetta, 2008; Pitard et al., 2001; Shapiro et al., 2000), the risk increased with increasing duration of cigar smoking. For example, Pitard et al. found the OR for BC was 1.4 (95 percent CI: 0.8–2.6) for 1–29 years, 2.7 (95 percent CI: 1.3–5.7) for 30–39 years, and 3.8 (95 percent CI: 2.1–7.1) for ≥40 years of smoking among primary cigar smokers compared to never-smokers (p value for trend <0.001) (Pitard et al., 2001).

The risk of BC has been shown to increase with the number of cigars smoked per day (McCormack et al., 2010; NCI, 1998; Pitard et al., 2001). In CPS-I, this trend was observed among primary but not secondary cigar smokers (NCI, 1998). Compared to never-smokers, the IRRs for BC mortality among primary cigar smokers were 0.78 (95 percent CI: 0.29–1.71) for 1–2, 1.68 (95 percent CI: 0.77–3.18) for 3–4, and 2.03 (95 percent CI: 0.97–3.73) for ≥5 cigars per day. Among secondary cigar smokers, the IRRs for BC were 1.02 (95 percent CI: 0.20–2.97) for 1–2, 2.36 (95 percent CI: 0.76–5.50) for 3–4, and 0.32 (95 percent CI: 0.00–1.80) for ≥5 cigars per day (NCI, 1998). Pitard and colleagues found that, compared to never-smokers, the ORs for BC were 1.3 (95 percent CI: 0.4–4.0) for 0.1–1.5 and 1.9 (95 percent CI: 0.8–4.4) for >1.5 cigars per day (Pitard et al., 2001).

___________________

⁹ This study was updated in Corrigendum, 2019.

cigar and pipe smokers who also smoked cigarettes (as well as 16 cases and 73 controls who were cigar and/or pipe smokers who did not smoke cigarettes). These three studies all had small case numbers, and none found significant associations. However, all associations were positive, and the Hassan results came close to significance: the OR for cigar use among noncigarette smokers was 2.2 (95 percent CI: 0.99–4.7). A larger study by Bertuccio and colleagues pooled 11 case-control studies and was sufficiently powered to find a significant association of 1.62 (95 percent CI: 1.15–2.29) (Bertuccio et al., 2011). However, exposures were classified differently for cigar use among the various studies. For example, while all studies used primary cigar smoking, some had ever-smokers and some regular or current smokers. In a meta-analysis by Iodice et al. (2008), the risk of pancreatic cancer among cigar smokers was 1.53 (95 percent CI: 1.02–2.28). However, these studies showed evidence of heterogeneity, which suggests potential for bias. Ultimately, there is growing evidence for an association between cigars and pancreatic cancer.

Other Cancers Summary

Studies of the associations between cancers others than lung, head and neck, and oral cavity cancer have been limited by the small samples of cigar users in epidemiological studies and the corresponding relatively small numbers of cancer cases available for analysis. However, the available evidence suggest that cigar use is associated with increased risk of all cancers, smoking-related cancers, pancreatic cancer, and probably urinary tract cancers, including bladder, with higher risks with increased intensity or frequency of use and level of inhalation.

No information is available regarding risks by specific cigar type, with only one study reporting risks for cigars versus cigarillos. The literature search found no studies reporting specific risks for premium cigar users. However, based on the cigar literature, it is expected that daily or frequent long-term use of premium cigars would result in higher risk of these cancers.

Future studies need to evaluate the associations between cigar use and the incidence and/or mortality risks of pancreatic, esophageal, bladder, and urinary cancers, accounting for frequency of use, duration, intensity, cumulative exposure, and pattern of inhalation.

OTHER HEALTH CONDITIONS

The committee identified several studies that investigated lesser-known and understudied potential health effects: skin health (contact dermatitis), diabetes, eye health (lens opacification), hepatitis C, fertility,

and the health of cigar factory workers. Few studies examined primary cigar smoking. There appears to be some evidence that there are health effects for cigar factory workers.

Contact Dermatitis

Bonamonte and colleagues reviewed literature on tobacco exposure and contact dermatitis (Bonamonte et al., 2016). Despite a clear pattern between cigar manufacturing and contact dermatitis, the association with cigar smoking is unclear. The authors reported an association with smoking and contact dermatitis, but studies supporting this conclusion examined cigarettes rather than cigars. The connection to cigar smoking specifically and contact dermatitis is understudied, but it may exist, based on the association found with cigarette smoking.

Diabetes

The committee found no studies on exclusive cigar use and diabetes but identified one study on combined cigar/pipe smoking. While the role of cigars cannot be definitively determined using this combined cohort, the data are presented to indicate concern. Future studies need to examine exclusive cigar use and diabetes. Cigarette smoking is associated with an increased risk of type 2 diabetes. In a study by Wanamethee and colleagues, diabetes incidence was examined in primary cigar/pipe smokers (never-cigarette smokers who smoked cigars/pipes) and secondary cigar/pipe smokers (former cigarette smokers who smoked cigars/pipes) (Wannamethee et al., 2001). The reference group of never-smokers was defined as never-cigarette smokers who did not currently smoke cigars/pipes. They found no substantial association for primary pipe/cigar smokers. However, their reference group included individuals who were former cigar smokers, which may have attenuated a true positive significant association. They did find an association for secondary cigar/pipe smokers who switched from using cigarettes to pipe/cigars. Assessing the association between cigar smoking and diabetes is problematic, as cigar and pipe smoker were combined. Thus, more evidence is needed.

Eye Health

In a study of risk factors associated with lens opacification in Iceland, Arnarsson and colleagues found that pipe and/or cigar had a significant effect (OR: 2.5; 95 percent CI: 1.2–5.1) (Arnarsson et al., 2002). However, it is unclear how cigar smoking was defined, and it was combined with pipe use. Thus, nothing can be concluded regarding any possible association between cigar use and eye health.

those smoking other types of cigars because the majority of premium cigar smokers are nondaily or occasional users and because they are less likely to inhale the smoke.

Conclusion 5-6: There is strongly suggestive evidence that many of the health risks of daily exclusive cigar use in general (overall mortality; cardiovascular disease; lung, bladder, and head/neck cancer; chronic obstructive pulmonary disease; and periodontal disease) are significantly higher than those of never-smokers and lower than those of daily cigarette smokers.

Conclusion 5-7: There is moderately suggestive evidence that the health risks among primary cigar users in general (those who were never established cigarette users) are generally lower than among secondary cigar users (those who were former users of cigarettes) because secondary cigar users may be more likely to inhale the smoke. Likewise, concurrent users of premium cigars and other combustible tobacco products would experience greater health risks than those smoking only premium cigars.

Conclusion 5-8: There is insufficient evidence to draw conclusions on the health effects of premium cigars on

• Youth or young adults,
• Racialized and ethnic populations,
• Pregnancy,
• Those with underlying medical conditions,
• People with occupational exposures to premium cigars (e.g., cigar lounges, manufacturing), and
• Health effects compared to other cigar types.

DIFFERENCES BY SOCIODEMOGRAPHIC CHARACTERISTICS

Very few studies on the health effects of cigars examine different sociodemographic characteristics. Most studies did not examine differences by sex or were conducted only among male populations, given the relatively low rates of cigar use among women. Regarding overall mortality, two studies (Inoue-Choi et al., 2019; Lange et al., 1992) that included women considered differences by sex; in both studies, associations of cigar smoking with all-cause mortality were stronger among women. Among the studies that included nonwhite populations, most studies accounted for race and ethnicity as a confounder by either using it as a covariate in multivariable models or matching cases and controls based on race and ethnicity (see, for example, Hartge et al., 1985; Inoue-Choi et al., 2019; Malhotra et al., 2017; Morrison et al., 1984; Rodriguez et al.,

2010). No studies were identified that examined potential differences by race and ethnicity in the association of cigar smoking and health risks.

POTENTIAL HEALTH IMPLICATIONS OF FLAVORINGS IN PREMIUM CIGARS

In the research questions provided by FDA and the National Institutes of Health, the committee was asked what the impact of adding flavors to premium cigars would be (see also Chapter 2). Commonly marketed flavors in tobacco products, including cigars, are fruits/candy (e.g., grape, mango, melon, strawberry, apple, peach, berry), crème/butter, cinnamon, cheesecake, coffee/tea/chocolate, alcoholic beverages, and nonidentifi-able varieties (e.g., “tropical,” “cosmopolitan”). Some of the chemicals in these flavorings have known respiratory toxicity (e.g., diacetyl, cinnamaldehyde). Many of these chemicals are included on FDA’s list of additives shown to be “generally recognized as safe” under conditions of intended use (FDA, 2019); however, this designation applies to consumption and/or topical use, and the criteria do not include an examination of inhalation risks. In fact, the flavor chemical profile for flavored tobacco products is similar to that for candy (e.g., Swisher Sweet grape small cigars versus Kool-Aid grape mix), and flavored tobacco products may also have higher levels of some flavor ingredients per serving (Brown et al., 2014).

Different flavoring chemicals used in cigars may differentially influence toxicity in the production of oxidative stress, DNA damage, epithelial barrier dysfunction, and inflammatory responses with varying intensity and duration of exposure. Limited information is available on their adverse respiratory health effects even from manufacturer to manufacturer within the same class of flavorings. A significant concern exists regarding the ingredient purity and the general lack of oversight in manufacturing or marketing/communication (Kaur et al., 2018). Common flavorings have potential respiratory effects and have been shown to further enhance inhalation toxicity of tobacco smoke (Kaur et al., 2018; Paumgartten et al., 2017; Roemer et al., 2012).

Flavors in tobacco mask harsh taste, reduce throat irritation, and make smoke easier to inhale, which increases carcinogen exposure, nicotine intake, and addiction potential (Kostygina et al., 2016). Cigar flavors have been shown to increase the appeal of cigar smoking by masking the harshness and smell of tobacco (Delnevo et al., 2015). By doing so, flavorings may also make it easier for young and novice smokers to initiate tobacco use (King et al., 2013; Villanti et al., 2019, 2021). A study based in the Southeastern United States found that flavored cigarillos were mood-enhancing and flavors made cigar products more palatable (Sterling et al., 2015).

lems, including low birth weight and increased risk of sudden infant death syndrome (HHS, 2006; Vanker et al., 2017). The vast majority of SHS health effects studies are based on nonsmokers living with cigarette smokers who smoke in the home. Children are particularly susceptible to SHS harms. While the committee did not identify studies of disease risk in nonsmokers who live with cigar users who smoke indoors, the risks are likely to be similar to that of secondhand cigarette smoke for similar levels of exposure.

SHS consists of a combination of emissions related to passive burning of the tobacco (sidestream smoke) and exhaled mainstream smoke. For cigar smokers, exhaled smoke may contribute a smaller percentage to overall SHS due to small puff volumes in premium cigar smokers who do not inhale. The constituents of SHS from cigars are qualitatively similar to secondhand cigarette smoke. Particulate matter in smoke is an important toxicant, which contributes to CVD and pulmonary disease and possibly other diseases. Inhaled particulates cause oxidative stress and inflammation and affect autonomic nervous system function, which can promote disease.

A few studies have examined indoor air SHS from different types of cigars by measuring airborne particulates, CO, and PAH concentrations. The air concentrations depend on emission rates of particular substances from the cigar, the number of smokers, duration of smoking, size of the room, and ventilation. Studies generated SHS using smoking machines, volunteer smokers, or sampled real-life cigar social events. Large and premium cigars are smoked for much longer than cigarettes, so the peak concentrations of particles and CO are higher after smoking. However, if multiple cigarettes are smoked during the day, the cumulative exposure could be higher.

Particulate emissions in one small study averaged 0.2–0.7 mg/min for cigars compared to 0.7–0.9 mg/min for cigarettes (Klepeis et al., 2003). One premium cigar was tested in this study with an emission rate of 0.35 mg/min. Mass-normalized emissions were lower for cigars than cigarettes: 3.3–5.2 versus 7.0–7.6 mg/gram smoked. The premium cigar studied yielded 3.7 mg/g smoked. The particle size distribution was similar for cigarettes and cigars, with most of the particles between 0.02 and 2 micrometers. In another study, volunteers smoked one cigar (Italian Toscanello)¹⁰ over 30 minutes or three cigarettes, one per hour, in a test room with measurement of particle concentrations over 200 minutes (Protano et al., 2017). The peak particle concentration was substantially higher

___________________

¹⁰ This is likely a premium-like rather than premium cigar—it is not clear whether it is long filler (or shredded tobacco), and some might be machine made. This is a borderline product, like Acid cigars.

for cigar compared to cigarette smoking, and the cumulative predicted lung particle deposition for children was also higher with cigar smoking.

Klepeis et al. (1999) measured air concentrations of CO, particulates (PM_2.5), and PAH from different cigars in three environments. In a vacant office on different occasions, five different cigars were machine-smoked (for 7 to 40 minutes, depending on the cigar) with measurement of CO levels over time. The emissions included both mainstream and sidestream smoke. One premium cigar was studied (Todo El Mundo¹¹ or Ashton); it was smoked for 28 minutes and generated a peak CO concentration of 15 ppm, with an average emission rate of 42 mg/min or 82 mg/gram smoked. Other smaller cigars generally generated similar peak CO concentrations, while having higher emission rates per gram tobacco smoked. In a residence, volunteer smokers smoked a large cigar (Santona, 13.2 g;¹² Paul Garmirian, 15.4 g¹³) or a Marlboro cigarette. The cigars were smoked for 1.3 and 1.5 hours and compared to one cigarette. One cigar generated a peak CO concentration of 3 ppm, with an average emission rate of 14 mg/min or 130 mg/gram smoked. The other cigar generated a peak respirable suspended particle concentration of 0.35 mg/m³, with an average emission rate of 0.98 mg/min or 8.2 mg/gram smoked. The cigarette generated a peak respirable suspended particle concentration of 0.16 mg/m³, with an average emission rate of 1.9 mg/min or 43 mg/gram smoked. Finally, the investigators sampled CO concentrations in two large cigar social events with a high degree of ventilation due to open doors and windows. Indoor CO concentrations were 5–11 ppm, averaging around 6 ppm. The contribution of cigar smoking was similar to that measured on the freeway driving to the events.

Secondhand Smoke Summary and Conclusion

The limitations of this review are the small number of studies and that few premium cigars were tested. Nonetheless, it seems clear that concentrations of secondhand cigar smoke can be similar to or greater than that from cigarettes. The emission rates appear to be lower for cigars, but cigars are smoked for much longer periods. It is likely that the health effects of indoor premium cigar and cigarette smoking would be similar for a similar duration and intensity of exposure. These could include increased risk of heart attack, respiratory symptoms, more severe respiratory infections in adults, lung cancer, lower respiratory tract infection,

___________________

¹¹ The study says “equivalent for Ashton” and that “a Todo El Mundo cigar of similar size to the Ashton was used for measurement of physical characteristics.”

¹² Given this cigar’s obscurity and size, it was likely a premium cigar.

¹³ This cigar meets the committee’s definition of premium.

and otitis media in children. Evidence is lacking about the extent of secondhand exposure to premium cigar smoke. Of particular concern with respect to harms from secondhand smoke exposure are workers in venues where premium cigars are commonly smoked, such as cigar lounges.

Conclusion 5-10: There is sufficient evidence that premium cigars generate considerable levels of secondhand smoke; however, there are insufficient data on the health risks associated specifically with exposure to premium cigar secondhand smoke. It is plausible that since the constituents emitted from premium cigars are similar to constituents from other tobacco products, the health risk might be the same, but the extent of secondhand premium cigar exposure is unknown.

ADDICTION POTENTIAL

Background

Tobacco addiction is a pathological pattern of compulsive use despite negative consequences. Addiction involves a constellation of symptoms clinically referred to as “tobacco use disorder” or “tobacco dependence,” involving tolerance, withdrawal, craving, loss of control over use, neglect of other life activities, and others (APA, 2013; HHS, 2008). Addiction is believed to be the primary driver that maintains regular tobacco use, interferes with quitting, and perpetuates chronic use (HHS, 2008). Thus, the greater the inherent addictiveness of a tobacco product, the more likely users will experience high levels and durations of exposure to its toxins, and, in turn, risk of adverse health effects. In addition, the symptoms of tobacco use disorder interfere with quality of life and results in significant distress and impairment of social or occupational functioning (APA, 2013; Hughes, 2006), making it an important disease outcome.

Research Questions and Approach to Evidence Review

The research questions addressed in this section were (1) are premium cigars addictive and (2) is their addiction potential different from that of other cigar products and other noncigar tobacco products, and what is the reason for the presence or absence of cross-product differences in addictiveness? Because of the absence of direct empirical evidence for premium cigars, two approaches were taken. First, the committee examined biological plausibility. This involved reviewing the extent to which premium cigar characteristics relevant to addictiveness (e.g., nicotine delivery and sensory features) paralleled the profile of features known to make a tobacco product addictive. Second, the committee reviewed

empirical literature indicative of the addictiveness of nonpremium cigars and evaluated the certainty to which making inference generalizations from nonpremium cigars was possible. Newly calculated comparisons of tobacco dependence between premium cigar users and users of other tobacco products commissioned for this report were also reviewed (Jeon and Mok, 2022). The methodology to approaching evaluating the biological plausibility and indirect empirical literature follows.

Empirical Evidence and Biologic Plausibility on Addictiveness of Premium Cigars and Other Tobacco Products

Determining Biological Plausibility

Risk of tobacco dependence onset, duration, and severity is a function of the inherent addictiveness of the product and by the extent of exposure to it (i.e., chronicity, frequency, and quantity of use). That is, the extent to which increasing tobacco product use translates into successive increases in risk of addiction is augmented for products with high addictiveness. Nicotine is the principal addictive constituent in tobacco smoke that underlies tobacco dependence (Benowitz, 2010). Nicotine activates the brain’s reward system and other neurocircuitry, causing pleasure and desirable (i.e., reinforcing) effects (HHS, 2014). Chronic nicotine exposure causes neuroadaptations that underlie addiction and the dependence syndrome (HHS, 2014). Tobacco and nicotine products that produce rapid spikes in blood nicotine in the form of “boluses” that are delivered quickly to the brain have a high potential for addiction, including inhalable products with efficient pulmonary delivery (e.g., combustible cigarettes, e-cigarettes) (Benowitz, 2010). Products that deliver appreciable levels of nicotine via the oral mucosa at a slower rate with no pulmonary delivery can also be addictive (e.g., smokeless tobacco), albeit less so than conventional cigarettes. Therapeutic nicotine products that deliver nicotine more slowly via nonpulmonary routes are minimally addictive (e.g., Le Houezec, 2003).

Nicotine is necessary but not sufficient to cause addiction. Nicotine has direct reinforcing effects but also acts as a reinforcement-enhancer that augments the rewarding effects of nonpharmacological stimuli, including those associated with the tobacco self-administration sequence (e.g., taste, smell, sight of smoke clouds, hand-to-mouth movement, airway sensations) (Chaudhri et al., 2006). Repeated pharmacological exposure to nicotine in concert with these other pleasant cues synergistically increase the reinforcing effects of tobacco product use (Chaudhri et al., 2006). Thus, nicotine and tobacco products that provide more opportunities to provide high-intensity sensations and other pleasant stimuli in concert with self--

administration are likely to be more addictive (e.g., inhalable or flavored products). Tobacco products involving hand-to-mouth movements that can provide pleasant tastes and other oral sensations without stimulating the airways are also addictive (e.g., smokeless tobacco), albeit to a lesser extent than inhalable products. By contrast, nicotine products with very few stimulus opportunities (e.g., transdermal patch) are minimally addictive.

Given this conceptual premise, any tobacco product that delivers higher amounts of nicotine to the blood and does so quickly and in concert with a greater variety of pleasant sensations is likely to have a higher addiction potential. Therefore, the committee’s review of biological plausibility involved examining features of premium cigars likely to indicate the impact of their blood nicotine and sensory stimuli. The committee also integrated studies of the effect of nonpremium large cigars and other cigars on nicotine yield with this. Inferences regarding whether premium cigars are addictive involved considering comparisons of nicotine yield and sensory profile to other tobacco products with known addictiveness.

Evaluating Experimental Research on Addiction Potential

Abuse liability/addiction potential assessment refers to a host of experimental and quasi-experimental research paradigms designed to examine the effect of exposure to a tobacco product on intermediate end points that are proxy outcomes indicative of addictiveness and likely to correlate with risk of dependence (Carter et al., 2009). These studies can involve controlled exposure to a certain “dosage” of a cigar (e.g., number of puffs) or restricted duration of ad libitum use (e.g., up to 10 minutes with as many puffs as desired). The outcomes include measures of the product’s “abuse liability,” including the subjective pleasant effects, ability to suppress smoking urge or withdrawal symptoms (two elements of tobacco dependence), willingness to expend effort or money to obtain more of the product, and amount used under unconstrained conditions. Experimentally assigning participants to exposure to one cigar product versus another (or between cigars and other tobacco products) using randomized between-subject or within-subject crossover designs, allows for causal inferences about the relative abuse liability. These studies’ ecological validity is challenged by several factors, including the existing use preferences of the population tested (e.g., whether the sample includes nonsmokers of cigars) and possibility that the experimental product may differ from the participant’s preferred product. Abuse liability studies can provide inferences about whether a product is addictive based on use changes and pre- versus post-smoking outcome (e.g., withdrawal or urge). Differences in the effects of a particular cigar product relative

to other tobacco products with known addictiveness can also address whether it is addictive and its comparative addictiveness. If the effects of a cigar product on abuse liability indexes are similar to those caused by use of another tobacco product with known addictiveness, it can be inferred that the cigar is indeed addictive.

Evaluating Observational Epidemiologic Research on Tobacco Dependence

Observational research studies of dependence involve administering questionnaire or interview measures of tobacco dependence symptoms to populations of users of the product. A key metric of a product’s dependence potential is the overall prevalence of experiencing dependence, speed of acquisition of dependence symptoms, or mean number or severity of dependence symptoms in a population. To address the question of whether premium cigars are addictive, estimates of whether the prevalence or severity of tobacco dependence symptoms among users of premium cigars are nonnegligible (i.e., different from zero) were made. This included a review of estimates in other (nonpremium) cigars for comparison.

To address that question and also their comparative addiction potential to other products, dependence symptom metrics were compared between populations of cigar users and users of other tobacco products. This approach was also used to compare premium to nonpremium cigar users. Sampling and selection biases and poly-tobacco use are important considerations. These types of population-wide estimates are influenced by the overall frequency, quantity, and chronicity of use in the respective population. Use levels are due to the product’s inherent dependence potential (i.e., addictiveness) but also many other factors (e.g., price, ease of access, marketing, cultural trends). Some cigars might be more difficult for users to access due to their higher cost or due to lower accessibility (e.g., sold only in specialty shops). Because individuals might not be able to access certain cigar types on a regular basis, their ability to develop dependence symptoms from their use may be lower. By contrast, combustible cigarettes and certain types of mass-market cigars might be more widely available and provide ample opportunity for individuals to become frequent users and develop dependence symptoms. For these reasons, cross-population comparisons in overall prevalence or severity of dependence symptoms between users of premium cigar versus other tobacco products provide only a proxy for differences in the addiction potential between premium cigars and other tobacco products.

Other observational methods involve examining the association of frequency, quantity, and chronicity of use with dependence risk or severity. The magnitude of association between product exposure and dependence

may also provide an estimate of its addiction potential. This approach is likely to provide a stronger estimate of inherent addictiveness, over and above factors that contribute to variation in exposure. However, reverse causality and criterion contamination (some measures of dependence include frequency of use as symptom indicator) affect these designs, necessitating scrutiny of temporal precedence and the outcome construct.

An important consideration is that epidemiologic studies need to be limited to exclusive users of a single tobacco product so that the tobacco dependence symptoms reported by the user can be ascribed to that product. Tobacco dependence symptoms among poly-tobacco product users cannot be differentiated to a specific product. This approach reduces the generalizability of the results to the overall population of users, which includes high proportions of poly-tobacco product users (Kasza et al., 2017). In a PATH study, poly users are 2–3 times more likely than single-product users to report higher levels of nicotine dependence and could presumably have greater vulnerability to tobacco dependence than exclusive users (Strong et al., 2017). Consequently, the analyses may be underestimates of the level of dependence of premium cigar (and other tobacco product) users.

Results of the Evidence Review

Biologic Plausibility

Given the above considerations, the rate of blood nicotine delivery and extent of pleasant sensory cues experienced during self-administration of premium cigars will provide information regarding their dependence potential. As noted in Chapter 2, premium cigar smoke emissions from a puffing machine contain nicotine at levels that appear to be at least equivalent to nicotine in smoke in nonpremium cigars and other combustible tobacco products (e.g., conventional cigarettes) (Fant and Henningfield, 1998), although direct comparisons are complicated because of differences in the methodologies used to create machine-generated puffs of different cigars and cigarette products. An important consideration for nicotine effects is the rate of systemic absorption. This is influenced by the pH of the smoke, the extent of inhalation into the lungs, and the pattern of puffing. At lower pH, the nicotine molecule is more highly protonated, while at higher pH, it exists more in the unprotonated form. Unprotonated nicotine permeates cell membranes more easily, so smoke with alkaline pH facilitates nicotine absorption across the oral mucosa. Premium cigar smokers anecdotally report holding cigar smoke in their mouths, which could be a source of oral nicotine absorption. As discussed in Chapter 2, large cigars, including premium cigars, tend to be more

alkaline than combustible cigarettes (Fant and Henningfield, 1998; Henningfield et al., 1999; NCI, 1998).

Conventional cigarette smokers are more likely to inhale and to inhale more smoke more deeply than cigar smokers do, particularly cigar smokers with no history of conventional cigarette smoking (Fant and Henningfield, 1998; NCI, 1998; Wald and Watt, 1997). However, studies involving objective indexes of inhalation (e.g., CO; lung imaging) suggest users of large cigars do inhale smoke, including noncigarette smokers and individuals who self-report not inhaling (Claus et al., 2018; McDonald et al., 2002; Pickworth et al., 2017b; Rosenberry et al., 2018). Objective inhalation exposure studies suggest large cigars exhibit inhalational exposure that is at least equivalent to other cigar products. Research on the nicotine yield of large cigars indicates blood nicotine boosts similar in magnitude to conventional cigarettes and sometimes larger than other cigar products (Claus et al., 2018; Pickworth et al., 2017b; Rosenberry et al., 2018). Despite no rigorous research on inhalation patterns or nicotine yield from premium cigars, one would expect similar inhalation and nicotine yield relative to nonpremium large cigars. No specific studies exist on the pharmacokinetics of blood nicotine from premium cigar use.

Premium cigars, like other cigar products, provide the sensations and stimuli shown to be important to the dependence potential of tobacco products (e.g., hand-to-mouth movements, taste, smells, airway sensations). They do not have characterizing flavors, per this report’s definition of the product class (see Chapter 1), and flavors are known to increase the addictiveness of other tobacco products (e.g., menthol-flavored cigarettes, non-tobacco-flavored e-cigarettes) (see Chapter 3; Goldenson et al., 2019; Wickham, 2015). One nationally representative study found that premium cigar smokers were less likely to report that “they come in flavors I like” than users of other cigar types, although close to half (48.6 percent) of premium cigar users reported this as a reason they use cigars (Corey et al., 2018). However, premium cigars do have distinctive tastes related to tobacco blends, curing processes, and sometimes infusions with various volatile chemicals, as discussed earlier in this chapter. Given these results and the similar features of premium cigars to other tobacco products that are addictive, there is reason to believe that premium cigars have sensory aspects to contribute to their addiction potential. Given the absence of added characterizing flavors in premium cigars, their sensory profile may not be equivalent to nonpremium cigar and other noncigar products that are available in characterizing flavors. Their absence of flavors could reduce the sensory-related addictiveness versus explicitly flavored products.

In summary, some research indicates that nonpremium cigars, in particular large cigars that are similar in size and other characteristics

(no filter), might have nicotine levels similar to other cigar products and potentially conventional cigarettes. Furthermore, premium cigar and nonflavored cigar products may have similar sets of sensorimotor characteristics that contribute to addiction potential. For these reasons, it is biologically plausible that premium cigars can be addiction promoting, provided the user has sufficient extent of level exposure (i.e., chronicity x frequency x quantity of use represents the totality of exposure).

Experimental Addiction Potential Assessment of Cigars

The literature search identified seven total abuse liability studies of cigars, three of which examined large cigars and none of which examined premium cigars separately.

Large cigar addiction potential studies

Claus et al. (2018) studied adult, exclusive cigar users (N = 77) who smoked their own-brand product ad libitum for up to 1 hour. In the overall sample, smoking urge and withdrawal symptoms were significantly reduced by smoking, and the magnitude of tobacco withdrawal and urge suppression did not differ across people who used small cigars versus cigarillos versus large cigars. A group of four articles used partially overlapping samples of dual users of conventional cigarettes and cigars. Each of these articles applied the similar study design of ad libitum smoking of either a single experimenter-provided unflavored cigar or their own-brand cigarettes (Koszowski et al., 2015; Pickworth et al., 2017a,b; Rosenberry et al., 2018). Two of them studied large cigars. Pickworth et al. (2017b) studied dual users of cigarettes and cigars and made comparisons across groups who typically smoked either large or small cigars or cigarillos. After ad libitum smoking of a respective study product from the cigar class they typically smoked (Phillies Blunt [large cigar], Black & Mild [cigarillo], Winchesters [little cigar]), smoking urge was reduced from pre- to post-smoking for each cigar group. Ratings of product appeal and sensory effects had no group differences, although little cigars were rated significantly lower than cigarillos and large cigars on satisfaction. Rosenberry et al. (2018) studied dual users of cigarettes and large cigars (n = 17, 94 percent men, 77 percent African American) who smoked ad libitum either their usual cigarette brand or a study-provided large cigar (Phillies Blunt) in two laboratory sessions using the same design as above. Smoking cigarettes and large cigars each significantly reduced the urge to smoke from pre- to post-smoking. The cigars and cigarettes had no significant differences in the magnitude of urge suppression or subjective product appeal or satisfaction.

dependence indicators (Apelberg et al., 2014; Gomez et al., 2020; Odani et al., 2020; Rostron et al., 2016; Strong et al., 2017; Sung et al., 2018). One study also found that U.S. adult cigar users who reported using a cigar product with flavors had higher odds of early morning smoking, which is an index of dependence, than those who did not (Odani et al., 2020). One reason for the mixed findings could be because the types of cigars that the user groups smoked varied in each study (e.g., large or little filtered cigars). Users of certain types of cigars might have higher dependence levels, given variation in use frequency and inhalation patterns across different cigar types.

Several studies found that U.S. poly users of cigars and other tobacco products have higher prevalence and severity of dependence compared to exclusive cigar users (Rostron et al., 2016; Strong et al., 2015; Sung et al., 2018). None of these studies differentiated dependence symptoms associated with use of different tobacco products (e.g., asked about symptoms related to tobacco use more generally).

In this report’s commissioned analysis of PATH, exclusive premium cigar users had a prevalence and mean severity of tobacco dependence symptoms that was substantially lower than that of exclusive cigarette or smokeless tobacco users across waves (see Table 5-1) (Jeon and Mok, 2022) but comparable to that of exclusive hookah users.

Differences in mean dependence across different types of cigar products

The only published comparison of dependence symptoms across users of different cigar types was in a convenience sample of 77 participants in a clinical research study (Claus et al., 2018): large cigar smokers reported lower levels of nicotine dependence and baseline nicotine withdrawal than little cigar smokers but did not differ from cigarillo smokers (Claus et al., 2018). In this report’s commissioned PATH analysis, exclusive premium cigar users had a substantially lower prevalence and mean severity of tobacco dependence symptoms than exclusive filtered cigar users (see Table 5-1) (Jeon and Mok, 2022). Exclusive premium cigar users had dependence symptom prevalence and severity that was moderately lower than that of exclusive cigarillo users and slightly lower than that of exclusive nonpremium traditional cigar users.

As described in Chapter 3, premium cigar smokers smoke at lower frequencies and quantities than users of other cigars and cigarettes. None of the reviewed studies has comprehensively adjusted for differences in use frequency, use quantity, and other possible factors that may affect dependence vulnerability (e.g., mental health) (Dierker and Donny, 2008). Consequently, differences in dependence between users of premium cigars, nonpremium cigars, and other tobacco products may be due to one or more of these factors. Hence, it is difficult to rule out the possibility that

the substantial differences in use frequency of cigarettes and premium cigars are entirely driven by external factors rather than the product’s inherent addictiveness (e.g., availability, price, culture).

Association between use exposure and dependence of cigars

Cross-sectional studies have found significant associations of frequency of cigar use with prevalence and severity of dependence in U.S. youth and adults (Gomez et al., 2020; Strong et al., 2017). In a study of youth, the association of frequency of cigar use and dependence was weaker than the association of cigarette use frequency and dependence (Gomez et al., 2020). However, in a study of adults, the magnitude of association of daily compared to nondaily use with dependence for cigar use was not different from the association of daily use and dependence for cigarettes (Strong et al., 2017). No longitudinal studies of cigars assessed use frequency prior to dependence outcomes.

The commissioned PATH analysis calculated the associations between the level of tobacco product use exposure (assessed as past 30-day use frequency) and severity of tobacco dependence for each product using the 0–100 scale (see Table 5-2) (Jeon and Mok, 2022). For Waves 1–3, the strength of association of multiple indicators of past 30-day use frequency (daily versus nondaily, number of days, and 6+ versus <6 days) with dependence was comparable between exclusive premium cigar users and exclusive hookah users, and for some exposure metrics, it was comparable between exclusive premium cigar users and exclusive users of other noncigarette tobacco products. For Waves 4–5, the difference in dependence severity between users of <6 days versus 6+ days in the past 30 days was also comparable between exclusive premium cigar users and exclusive users of other products (except for cigarettes and smokeless tobacco, which produced stronger associations; see Table 5-2). For Waves 4–5, associations of number of days used and daily versus nondaily use with dependence severity were not statistically significant and produced wide confidence intervals, potentially due to the small number of daily exclusive premium cigar users (Jeon and Mok, 2022). However, in general, all waves had few daily exclusive cigar users and exclusive hookah users, suggesting the need for caution in interpreting this result. For all products, the number of exclusive users in <6 days versus 6+ days groups was sufficient to produce reliable estimates. For that comparison, the data suggest significant associations between use frequency and dependence for each product; the magnitude of association for premium cigars was comparable to that of hookah and less than cigarettes and smokeless tobacco (Jeon and Mok, 2022).

In summary, the prevalence and severity of tobacco dependence among exclusive premium cigar users (and traditional cigar users, includ-

ing premium/nonpremium users) is modest. The extent of dependence in these populations is substantially lower than among exclusive users of cigarettes and several other addictive tobacco products but comparable to hookah users. The association of level of exposure and dependence severity similarly is substantially smaller than that for cigarettes and on par with that for hookah.

Addiction Potential Summary and Conclusion

Two studies show that users of premium or large cigars have low population-wide prevalence and severity of dependence symptoms but at levels that are not negligible. The prevalence evidence estimates only modestly altered the committee’s conclusion regarding the addiction potential of premium cigars because these estimates are confounded with ease of accessing premium cigars versus other tobacco products and preexisting dependence vulnerability. Premium cigars are more difficult to access than other tobacco products because of their high cost and limited availability (e.g., at specialty retailers). Consequently, only a subset of the population can become frequent users of premium cigars and have sufficient opportunity to develop dependence symptoms from their use, which might reduce the population-wide prevalence and severity of premium cigar dependence symptoms. Furthermore, in comparison to users

of other tobacco products, premium cigar users have disproportionately fewer pre-existing risk factors for nicotine dependence (e.g., low socioeconomic status, mental health problems, comorbid substance use; see Chapter 3). In addition to the population prevalence estimates, several well-designed abuse liability/addiction potential studies of large cigars with similar characteristics to premium cigars consistently show that abuse liability outcomes are significantly increased from pre- to post-cigar administration and that the effects of large cigar administration on abuse liability outcomes do not differ from other addictive tobacco products. A strong biological plausibility exists that premium cigars possess the features (i.e., rate/amount of nicotine delivery, pleasant stimuli) liable to make them as addictive as other tobacco products with known addiction potential (e.g., smokeless tobacco). Therefore, there is moderately suggestive evidence that premium cigars can be addictive.

One moderate-quality study shows that users of premium cigars have lower prevalence or severity of dependence symptoms than cigarette users; another low-quality study shows lower dependence in cigar users compared to cigarette users. One abuse liability study found no differences between large cigars and cigarettes, but studies of other (nonlarge) cigar products find lower abuse liability compared to cigarettes. Premium cigars possess features (i.e., rate/amount of nicotine delivery, pleasant stimuli) that make them less likely to be addictive than cigarettes, particularly cigarettes with characterizing flavors. Overall, there is moderately suggestive evidence that premium cigars are less addictive than cigarettes.

The commissioned PATH analyses showed that prevalence and severity of tobacco dependence among exclusive premium cigar users (and traditional cigar users, including premium/nonpremium users) is modest (Jeon and Mok, 2022). The extent of dependence in these populations is substantially lower than among exclusive users of cigarettes and several other addictive tobacco products but comparable to hookah users. The association of level of exposure and dependence severity similarly is substantially smaller than that for cigarettes and on par with that for hookah.

Gaps in the literature include the absence of longitudinal data on trajectories of dependence for premium cigar users to provide information on the speed of dependence acquisition, as is research that compares the prevalence and severity of dependence across different demographic groups, which can provide insight into the role of premium cigars in health disparities. Research with detailed assessment of premium and nonpremium cigar use patterns, including inhalation behaviors, in epidemiologic studies of dependence would provide useful information about the inherent addictiveness of cigar products. Controlled abuse liability

studies of premium and nonpremium cigars with different product characteristics (e.g., flavors, pH, size) is lacking and would be useful for informing regulation of cigar products, particularly research isolating the impact of flavors.

Conclusion 5-11: There is moderately suggestive evidence to support the biological plausibility that regular cigar smoking in general can be addictive. It is likely that this is also true for premium cigar smoking, based on nicotine delivery characteristics, abuse liability studies, and epidemiological data. The magnitude of premium cigar dependence appears to be less than that of cigarette smoking and smokeless tobacco use dependence. The extent of addiction is likely to depend on the patterns of use.

RESEARCH GAPS

As noted in the chapter opening, premium cigars’ potential adverse health effects need to be viewed in the context of harms of combusted tobacco smoking broadly, and the mechanisms of tobacco smoke toxicity and biomarkers of toxicant exposure are applicable to understanding the potential harms of premium cigars. However, the committee identified key research gaps in the health effects literature for premium cigars specifically (see Box 5-2) and provides two recommendations to address these gaps (see Recommendations 4 and 8 in Chapter 6).