METHODS

Sample

Since the implementation of the Massachusetts Tobacco Control Program in 1993, The Center for Survey Research at the University of Massachusetts, Boston has been collecting data on the evolving tobacco related attitudes and behaviours of adults and teenagers across the state. The Massachusetts Tobacco Survey (MTS), conducted in 1993, supplied baseline data for both youth and adults on their prevailing beliefs and practices at the programme's initiation. The Massachusetts Adult Tobacco Survey (MATS) began in 1995 and continued through to December 2000, surveying approximately 225 adults each month on various programme and policy related topics. Both the MTS and MATS are random digit dial telephone surveys that feature two distinct components. The first, a brief interview with any one adult in the household, enumerates the adults in each household and provides demographic and smoking status information for each adult. From this interview, one adult is then randomly selected to be interviewed in depth. The MTS employed an additional oversampling strategy, whereby smokers and minorities were oversampled, while MATS did not employ the oversampling procedure. Response rates for MTS were 78% for the household interview and 78% for the in-depth interview. For MATS, response rates for the household interview have ranged from 68–76% and for the in-depth interview have been between 79–81%.^6,7

Measures

Cigar smoking prevalence and experimentation

Both the MTS in 1993 and the ongoing MATS included a question designed to estimate the smoking prevalence of non-cigarette forms of tobacco. The question mirrored that used to determine current cigarette smoking prevalence and was worded as follows: “Do you now smoke a pipe, cigar, or cigarillos, every day, some days, or not at all?” In order to focus solely on the prevalence of cigar and cigarillo smoking, in December of 1996 a question was added to help distinguish these from pipe smoking: “Which of these tobacco products do you use?” Using this question, estimated proportions of pipe smoking (according to sex and age) were removed retrospectively from data that was collected before the introduction of the new question. These estimated proportions of pipe smokers were compared across the years for which there existed data (1997–2000) in order to determine the stability of the estimates. Proportions of pipe smokers in each demographic group were deemed very stable across these years of data collection, lending support to the rationale for applying them to data collected in prior years. Beginning in December of 1996, a question was inserted before the questions on current cigar smoking: “In the past 12 months, have you tried smoking a cigar—even a puff?” Respondents answering “yes” to this question were classified as having at least experimented with cigar smoking in the prior year.

Cigarette smoking prevalence

Determination of cigarette smoking prevalence followed established criteria: current smokers had smoked at least 100 cigarettes in their lifetime and smoked either “every day” or “some days” at the time of the survey; former smokers had also smoked at least 100 cigarettes in their lifetime, but smoked “not at all” at the time of the survey; and never smokers were those who had not smoked 100 cigarettes in their lifetime. This measure of smoking prevalence was used as a comparison characteristic of those who smoked cigars.

Perceptions of health risks

A measure of the perceptions of health risks was obtained by asking respondents, “Do you believe that switching from cigarettes to cigars reduces a smoker's chance of illness?” This question was also added in December of 1996 and was asked of all respondents.

Data analysis

All data were weighted and analyses were conducted using SUDAAN,⁸ in order to account for the complex stratified sampling design. Trends in cigar smoking were assessed using combined sample years, for the purpose of comparing similarly sized groups.

RESULTS

Trends in cigar smoking

Table 1 depicts trends in cigar smoking across time, by demographic predictors sex and age. As shown, there occurred a significant upward trend in use by young men (ages 18–34 years) between the years of 1993 and 1998; in fact, there was a tripling in the rate of current use over the course of six years (from 5.8% in 1993 to 18.2% in 1997-98). Following the 1997-98 period, however, there was a decrease, though not significant, in cigar use among young men (13.5% for the period between 1999 and the first half of 2000). In contrast to patterns among young men, cigar smoking prevalence for women was consistently very low and virtually unchanged across the years in question. Similarly, rates of usage for older men were found to be more stable and substantially lower than those for younger men. Aside from the increase in cigar usage by young men between 1993 and 1997-98, there were no other significant differences noted. It is also important to mention that the majority of the young men using cigars reported smoking them “some days” (98%) rather than “every day” (2%).

Cigarette smoking status of young male cigar smokers

Table 2 compares across time the cigarette smoking status of young, male cigar smokers. The percentage of cigar smokers who had never been cigarette smokers increased (from 23.3% in 1993 to 55.2% in 1999-2000), while the percentage of cigar smokers who were former cigarette smokers declined slightly (from 17.1% in 1993 to 9.4% in 1999-2000). Because of large confidence intervals, the differences between these estimates were not significant. The findings suggest that cigars were not being used increasingly as a substitute for cigarettes among those who had quit.

Experimentation

Results regarding levels of experimentation with cigars, by sex and age, are shown in table 3. Proportions were compared across time. Once again, young men appear to have been the primary experimenters, followed by their older counterparts. While slightly more than one in 10 young women admitted to experimenting with cigars, fewer older women reported this type of experimentation. In comparing trends across time, it appears that experimentation with cigars may be on the decline among young males (42.4% in 1997-98 v 35.6% in 1999-2000), but this change was not significant.

Perception of risk

Data collected between 1996 and 1999 were analysed by cigar smoking status, and results indicated a significant difference between the beliefs of cigar smokers and cigar non-smokers, with the former being three times as likely to believe cigars are a safer alternative to cigarettes (32.3% v 11.2%) (p < 0.05). Perceptions among cigar smokers were additionally examined by cigarette smoking status. Former cigarette smokers were shown to be much more likely to believe that cigars are a safer alternative to cigarettes (47.9%) than current cigarette smokers (22.9%) or those who have never smoked cigarettes (28.8%), though these differences are not significant at the 0.05 level.

DISCUSSION

Our results indicate that the recent glamorisation of cigars corresponds with an increase in both experimentation with, and current usage of, cigar products by young men in Massachusetts. The trend may have peaked in 1997-98. Current use of cigars rose from 5.8% in 1993 to 18.2% in 1997-98 among this group, constituting a highly significant increase (p = 0.001). Data collected in California in the years 1990 and 1996 show a similarly large increase in the proportions of young men currently using cigars during this time period.⁹ These increases have also been seen nationally among young men.¹ Interestingly though, while marketing has appeared to focus on both young men and women, women seemed to be more resistant to the lure of current cigar smoking. Though data show a small proportion of 18–34 year old women had experimented with cigars, these women did not display the same patterns of current usage that characterised men of the same age group. Fortunately, data collected since 1998 suggest that the explosive increase in cigar use among young men may be slowing, with prevalence figures for this group down to 13.5% for the period between 1999 and the first half of 2000. In addition, data on per capita consumption of cigars show a decrease for the entire US population, as well as for adult males, between 1999 and 2000.¹⁰

We thought it important to assess whether the cigarette smoking status of the young male cigar smoker had changed over time, as their numbers increased. Whether the new cigar smoker was a current, former, or never smoker of cigarettes is a variable crucial to understanding the nature of the cigar phenomenon. Analysis of the cigarette smoking status of the young male cigar smoker over time revealed that more of the new cigar smokers were those who had never smoked cigarettes (23.3% of the cigar smokers in 1993 v 55.2% of the cigar smokers in 1999-2000). In addition, a slight decline was revealed in the proportion of cigar smokers who were former cigarette smokers; an important finding that suggests cigarette smokers are not overwhelmingly turning to cigars as a substitute for cigarettes. Although the statistical power of these analyses is quite low because of small samples sizes of young male cigar smokers, the estimates are suggestive of trends for this population.

Though former cigarette smokers do not appear to be turning to cigars in increasing numbers, this situation does occur. An important question here concerns the potential for cigar smokers to view cigar smoking as a “safe” alternative to cigarettes. Since former or current cigarette smokers who smoke cigars have been shown to inhale more deeply than do those who have never smoked cigarettes,¹ and the health risks have been shown to be much greater with deeper inhalation, perceptions by cigar smokers of the safety of their cigar smoking are an important concern. Indeed, our results show cigar smokers to be three times as likely as those who do not smoke cigars to believe that switching from cigarettes to cigars reduces a smoker's chance of illness (32.3% v 11.2%), with former cigarette smokers the most likely among cigar smokers to believe that cigars are a safer alternative (47.9%). It will be important to continue monitoring trends in marketing and advertising of cigars over the next few years. This will allow for further examination of the links between activities of the cigar industry and population smoking patterns.