Over the years the tobacco and gaming industries have found common ground in their efforts to fight smoke-free policies.1^–3 Both industries fear that smoking bans will result in a loss of customers. The tobacco industry fears the loss of customers and sales from the impact that smoke-free policies have on the social acceptability of smoking, which ultimately results in a reduction of cigarette consumption.4 5 The gaming industry fears that smoking bans would drive away their smoking customers and/or reduce the time that smoking customers spend at the gaming venues.6 7 This conclusion by the gaming industry has been reinforced by the published reports of a relation between problem gambling and smoking.6 8^–10 This area of common ground has forged a powerful alliance between the two industries which have worked together to defeat smoke-free policies worldwide.1

Despite the recent proliferation of legislation for workplace smoking bans throughout the United States and worldwide, exemptions continue to be carved out for casinos and other gaming venues. In the United States alone, 22 states have passed smoke-free workplace laws which include bars and restaurants.11 Many states, in which gaming is legal, however, have not been successful in including gaming workplace sites in their smoke-free policies. This is primarily because of the heavy lobbying of the gaming industry. A key component in the argument against these restrictions is that a great many gamblers smoke and thus any restrictions would cause disproportionate economic hardship to this industry. In making this case, the gaming industry has been rather cavalier in its assessment of the number of its clients who smoke. In Nevada and in other states considering anti-smoking legislation, gaming lobbyists have routinely stated that more than 50% of their customers smoke.7 Casino owners and managers have claimed that between 50% and 90% of their customers smoke.7

Studies have also used these high smoking rates in developing economic impact statements on smoking bans on casino revenues. PricewaterhouseCoopers recently released a report for the Casino Association of New Jersey entitled “Smoking Ban Economic Effect Analysis.”12 In this report they presented their findings on the economic impact of a smoking ban on gaming in the state of Delaware and then translated those findings into a projected economic impact on New Jersey if a smoking ban were enacted that included Atlantic City, NJ gaming. In developing this report, they assumed a 50% smoking rate for the gamblers in both Delaware and New Jersey with no supporting documentation for these figures. These figures are considerably higher than those most recently reported for the US adult smoking rate (20.9%).13 A thorough search of the literature did not find any scientific study confirming this unusually high percentage of smokers among people who gamble in the United States, as claimed by the casino industry. Some of the figures used by these gaming lobbyists in fact originate from anecdotal information obtained in a survey of casino executives.7 With such a disparity between the claims of the gaming industry and the United States and worldwide smoking rates, it seemed important to conduct a thorough study to determine the accuracy of these figures. This project was designed to determine the percentage of gamblers who smoke while gambling in Nevada casinos, which have no legislatively mandated smoking restrictions in their gaming areas. Selection of three major tourist gaming destinations in Nevada including Las Vegas, arguably the gaming capital of the world, would serve as a good benchmark for other international gaming destinations.

METHODS

The overall goal of this study was to determine the percentage of gamblers who are smokers in Nevada casinos. To date, no scientific studies have been conducted to make that determination. In this study, we determined the percentage of gamblers who smoke by a method that actually counts the number of smokers in a room. A methodology for the determination of smokers in a room was published in 1980 by Repace and Lowry.14 Their method assumes that smoking in a large group is a random process and that habitual smokers on average smoke two cigarettes per hour and that each cigarette takes on average 10 minutes to smoke.15 16 The number of people then actively smoking in a room, at any one time, is one-third of the number of smokers in the room.14 This is a comprehensive study in that it looks not only at the percentage of gamblers who smoke in the three tourist destinations in Nevada but also by the type of game. It divides the data into those who play at the slot machines and those who play the table games (both low and high minimum wager tables).

Between 1 August and 27 October 2006 data were collected by teams of two people with tally counters. A team would walk through a casino with one person counting the number of gamblers and the other person counting the number of gamblers who were smoking. The team would cover the entire casino, walking up and down all of the slot areas as well as the gaming tables. For the purposes of this study table games included craps, roulette and all card games with the exception of the poker rooms. Players in the poker rooms were not included because many of the poker rooms in Las Vegas, Reno/Sparks and Lake Tahoe are now smoke-free and thus would have led to a systematic underestimation of smokers. People seated at bars in the casinos were also not counted. The study results are limited to those gamblers on the casino floor. Separate counts were recorded for those people playing at table games with $25 minimum play and those playing at tables with a less than $25 minimum play in order to determine whether any significant differences would be observed for the higher end players. We visited each casino twice, once in the afternoon and once at night, in order to get multiple counts at each gaming venue.

Eight casinos on the strip were randomly selected to be surveyed. They included the Bellagio, Caesar’s Palace, Excalibur, Luxor, Mandalay Bay, New York, New York, Venetian and Wynn’s. These are all major resorts on the Las Vegas Strip and should provide a good representation of the clientele gambling on the Las Vegas Strip. Data were collected from seven of the major casinos in the Reno/Sparks area. The Reno/Sparks area casinos surveyed were the Atlantis, Boomtown, Circus Circus, Grand Sierra Resort, John Ascuaga’s Nugget, Peppermill and the Silver Legacy. Data were also collected from three major South Shore, Lake Tahoe casinos. The casinos surveyed were Harrah’s, Harvey’s and MontBleu.

Statistical testing for the difference between two proportions was used.17 The null hypothesis was that the proportions (percentages) between the independent samples were equal. Estimated standard error and 95% margin of error were calculated. For each Nevada location, mean, standard error of mean and margin of error for sample average were determined. ANOVA was used to test the difference between means. Significance level was set at p<0.05. Unless otherwise noted, statistical data are reported as percentage (95% CI).

RESULTS

Data were collected from three different tourist destination areas within the state of Nevada in order to obtain a more generalised representation of the percentage of gamblers who smoke throughout the state. We collected data from eight Las Vegas Strip casinos, seven Reno area casinos and three Lake Tahoe casinos. Mean percentage of smokers between locations did not significantly differ (p = 0.43) as determined by ANOVA. The results from each of these areas are presented below.

Las Vegas casinos

Data were collected over a three-day period in August from casinos located on the Las Vegas Strip. We observed a total of 7633 gamblers, among whom 516 were smoking, for a calculated percentage of 20.3% (95% CI 0.9%) smokers (table 1). The percentage of gamblers observed smoking was not significantly different from the reported percentage of Americans who smoke (20.9% (95% CI 0.6%)) (p = 0.19).

The results of the study are shown in table 1. Each value is a total from two separate surveys of the casino. Overall, we see that 7633 total gamblers were counted with 516 of them being smokers. This results in a 6.76% smoking rate and then multiplying that by three gives us a total of 20.3% smokers among the gamblers on the strip. The percentage of smokers in these various casinos ranged from a low 16% in the Bellagio to a high of 24.3% at the New York, New York. Interestingly, the percentage of smokers among the combined table games was significantly lower, at 17.9% compared to 21% of slot machine players who were smokers (p = 0.0003).

Reno/Sparks casinos

The results of the study are shown in table 2. Overall, 4737 gamblers were observed together with 339 smokers for a calculated percentage of 21.5% (95% CI 1.2%) of smokers. The percentage of smokers did not significantly differ from the US population (20.9% (95% CI 0.6%)) (p = 0.32). As with the Las Vegas results, significantly fewer smokers were observed at the gaming tables, 14.9%, than at the slot machines, 23.0% (p<0.0001).

Lake Tahoe casinos

The data for the Lake Tahoe casinos are presented in table 3. Overall, 1682 gamblers were observed and 92 of them were smokers, for a calculated percentage of 16.4% of smokers. The same trends were observed in these casinos as were seen in the Las Vegas and Reno/Sparks casinos with fewer smokers observed at the tables (13.2%) than the slots (17.7%). As with the Reno/Sparks casinos, the percentage of smokers observed at the “high” minimum wager table games at the Lake Tahoe casinos is based upon only a few observations.

Percentage of gamblers by type of game

The data collect in tables 1–3 are presented in a different way in order to show the overall results for each of the games we surveyed (table 4). The results are presented as the percentage of gamblers at the table games (high and low minimum wager tables) and slot players for all observations made in this study. For all table games 176 smokers were observed from a total of 3222 gamblers for a calculated percentage of 16.4% smokers. When separated into high and low table games the percentages of smokers were calculated at 24.9% and 15.3%, respectively. The calculated percentage of smokers among the slot machine players was 21.3%.

DISCUSSION

In this study we observed gamblers and smokers at three major tourist destination gaming centres in Nevada. Overall we counted 14 052 gamblers and observed that 947 of them were smoking. Using the method of Repace and Lowry14 to determine the number of smokers in a room, we determined that 20.2% (95% CI 0.7%) of the gamblers observed in our study were smokers. These numbers are statistically similar to the results recently published,13 showing that in 2005 approximately 20.9% (95% CI 0.6%) of Americans smoke (p<0.0001). It is important to note here that the percentage of gamblers who smoke would be even smaller had this study included those gamblers in the casino poker rooms (which are almost all non-smoking facilities). We also stratified our data for all locations by high gaming tables, low gaming tables and slot machines. The overall percentages of smokers at each type of gambling activity were 24.9%, 15.3% and 21.3%, respectively (table 4). Significantly fewer smokers were observed between table games and slot machine players (p = 0.003 to <0.0001)

It is interesting to note the consistency in the percentage of smokers in the various casinos and at the three different locations in Nevada. We found no significant difference between mean percentage of smokers between locations. Twenty of the 21 casinos observed had a smoking rate of between 14.9% and 25.2% with nine casinos below 20% and 11 casinos above 20%. Furthermore, we found that the percentage of gamblers who smoke at Las Vegas and Reno/Sparks Nevada tourist destination casinos did not significantly differ from the overall percentage of smokers in the United States (p = 0.18 and 0.32). The percentage of gamblers who smoked at Lake Tahoe casinos was significantly lower than the national percentage of smokers (p<0.0001). One limitation to the study is that the data were not weighted in any way based on the actual number of gamblers at the various tourist destinations (Las Vegas strip, Reno/Sparks and Lake Tahoe). In addition, the number of casinos sampled, (by percentage) in Las Vegas was smaller in comparison to Reno/Sparks and Lake Tahoe owing to the greater number of major casinos on the Las Vegas Strip. Any overall determination of gamblers in Nevada, as a whole, would be dominated by the Las Vegas Strip casino results.

In conclusion, it appears that the percentage of gamblers who smoke in Nevada casinos is not significantly different from the overall percentage of smokers in the United States. This suggests that, overall, gamblers in Nevada casinos are not more likely to smoke than the general American population. Other worldwide gambling destinations might be likely to experience similar results to those we found in Nevada, with the percentage of gamblers who smoke at those casinos reflecting the smoking rate of the overall population for that area/country/region. The results of this study challenge past projected economic impact studies regarding smoking bans on gaming revenues that have used smoking rates greater than the overall general population smoking rate.