The main finding of this study is that exclusive male cigar smokers age 40 + years had significantly increased mortality only from chronic lower respiratory diseases (HR = 2.60, CI = 1.04–6.50), which was based on 6 deaths. We found no statistically significant evidence among exclusive cigar smokers of increased mortality from all causes, heart diseases, malignant neoplasms, cerebrovascular disease, smoking related diseases or other causes.
Our findings were similar to those from Inoue-Choi et al. (3). They used restricted NHIS-Linked Mortality Files and found no elevated mortality among exclusive current and former cigar smokers. Current daily cigar smokers had elevated mortality from cancer (HR = 2.27, CI = 1.23 to 4.19). However, the increase was not attributable to cancers of the oral cavity, esophagus, stomach, colorectal, or pancreas. Furthermore, nonsignificant increases in lung and bladder cancer were based on fewer than five deaths.
Another recent study by Christensen et al. (2) used data from the Tobacco Use Supplement to the Current Population Survey (TUS-CPS) linked to the National Longitudinal Mortality Study. It found that exclusive current cigar smokers ages 35–80 years had significantly elevated mortality for all causes (HR = 1.20, CI = 1.03–1.38), all tobacco-related cancers (HR = 1.61, CI = 1.11–2.32) and lung cancer (HR = 3.26, CI = 1.86–5.71). The authors observed that, although most cigar users did not smoke every day, the mortality increases were mainly due to daily users. No excess deaths were found among exclusive former cigar smokers.
Christensen et al. (2) and Inoue-Choi et al. (3) adjusted for only sociodemographic variables and survey years, whereas our models adjusted for more comprehensive confounders such as BMI, health status and geography. More importantly, those studies excluded use of other tobacco products, which eliminates 60% of all current cigar smokers from follow-up (8). Our approach includes the mortality experience of dual current cigar-cigarette users and current-former users of these two products. Our HRs for current and former cigarette smokers, regardless of their cigar use, were similar to those from previous studies by us (5) and others (2,9,10). Dual users had significantly elevated mortality risks for all outcomes except cerebrovascular disease. Former cigarette smokers who smoked cigars at the time of the survey had excess mortality for most causes, but the magnitudes were smaller than current cigarette smokers.
Other factors may be related to the low impact on mortality of cigar smoking in this study. Male cigar smokers over age 40 who never smoked cigarettes are more likely to be consumers of traditional cigars, which tend to be smoked less frequently and in smaller numbers than cigarillos or filtered cigars [10]. Additionally, differences in risk between cigar and cigarette smokers has been attributed for years to differences in inhalation practices (11,12,13). Although these factors might contribute to exclusive cigar smokers’ lower mortality compared to cigarette smokers, they do not seem to benefit former cigarette smokers who are current cigar smokers, who had elevated mortality for most diseases.
Our results also mirror those cited in a systematic review on cigars and health outcomes by Chang et al. (14). They defined primary cigar smokers as having had no history of cigarette use. The results were organized according to the number of cigars smoked daily, and two studies reporting all-cause mortality were cited (15,16). The first involved 15,000 primary cigar smokers in the American Cancer Society First Cancer Prevention Survey (15). Primary cigar smokers consuming 1–2 cigars daily had no increased mortality (1.02, 0.97–1.07), but those smoking 3–4 and 5 or more had elevated deaths from all causes (1.08, 1.02–1.15 and 1.17, 1.10–1.24 respectively). In the second study of 250,000 government-insured participants, most of whom were World War I veterans, fewer than 5 cigars per day was associated with no significant increase (1.04, 0.98–1.11) (16).
Similar results for other diseases related to primary smokers of 1–2 cigars per day were cited in the review by Chang et al. (14). For stomach, pancreas and bladder cancer, elevated risks were based on very small numbers of deaths and not statistically significant. Some cancer estimates were elevated, especially mouth/throat, esophagus, larynx and lung, but none were statistically significant. However, an older study by Shapiro et al. (17) using Cancer Prevention Study II (CPS-II) found that cigar smoking men age 30 + years had elevated mortality risk for lung, oral cavity/pharynx, larynx, and esophagus cancer.
Chang et al. (14) observed no increased mortality from coronary heart disease, stroke or emphysema among primary smokers of 1–2 cigars per day, but they did find an excess of deaths from aortic aneurysm (1.82, 1.11–2.81).
There are several limitations with our study. First, NHIS collects information on the use of cigarettes and cigars only once at survey enrollment for each participant. Second, we did not have information on the amount or duration of consumption among current smokers and the number of years since quitting among former smokers. Third, information on alcohol use and preexisting chronic conditions (e.g. diabetes, high blood pressure, high cholesterol), which are risk factors for premature death, was not available consistently after 1997 in NHIS, so we did not include them in our models. Similarly, there was no consistent data on physical activity and diet. Finally, we have limited statistical power and high standard errors due to the low prevalence of cigar use and small numbers of deaths for some outcomes.