Introduction

The words and pictures printed on the outside of cigarette packs have a significant impact on what people think (and feel) about what’s inside the packs. Unfortunately, even the numbers listed on cigarette labels can be seriously misleading. The effect of descriptors such as “light” and “mild” is even worse.

Since tobacco companies began marketing “light” and “mild” brands in the 1970s, market share for these brands has climbed steadily in Canada. Sales data indicate that brands labelled “light” or “mild” comprised approximately half of the total domestic market in 1994.

One result of this trend is that the number of brands commonly available to Canadian consumers has increased from approximately 80 in 1968 to just under 140 in 1996. Of these, 63% contain brand name descriptors such as “light,” “mild,” “smooth,” “extra light,” or “ultra mild.” In total, there are 24 quantitative (regular size, king size, 100 mm) and qualitative (light, mild, special, smooth, menthol) descriptors in use [see Table 1].

The health community has raised concerns that the introduction of more brands with “light” labels has succeeded in keeping in the market smokers who might otherwise have quit.

This concern is well-founded. Tobacco manufacturers acknowledged in the 1980s that they were losing smokers to health concerns, and investigated expanded product solutions as a vehicle for slowing the trend toward quitting:

It is no exaggeration to suggest that the tobacco industry is under siege. The smoker base is declining, primarily as a result of successful quitting. . . . Unmet needs of smokers that could be satisfied by new or modified products, products which could delay the quitting process, are pursued.(1)

There is evidence showing that this strategy gained some success shortly after the introduction of so-called lower tar and nicotine brands. For example, a 1978 industry document states: “we have evidence of virtually no quitting among smokers of those [”ultra” low tar] brands and there are indications that the advent of ultra low tar cigarettes has actually retained some potential quitters in the cigarette market by offering them a viable alternative.”(2)

Research of consumer attitudes and behaviour in the U.S. shows that many smokers perceive brands with lower tar and nicotine yields to be safer or healthier than “regular” brands. Higher quitting rates in the U.S. among smokers of higher tar cigarettes than smokers of lower tar cigarettes lends support to the theory that the presence of lower tar brands keeps some smokers in the market.(3)

The potential for these brands to delay or prevent cessation has not escaped tobacco company marketing strategies. Media plans for Medallion, marketed as a very low- tar brand, articulate a focus on advertising during New Year’s,(4) a time, of course, which is a popular target date for quitting.

Tobacco industry officials have testified under oath that they “have evidence . . . that certain smokers believe . . . that light, low tar and nicotine cigarettes are healthier than full-flavoured cigarettes.”(5) And a 1993 Gallup survey (6) found that half of all American adults believe that lower tar cigarettes are safer.

Unfortunately, smokers’ beliefs about improved health and safety from “switching down” are ill-founded. Researchers have written extensively on the discrepancy between reported levels of tar, nicotine and carbon monoxide derived from measurement systems in use in Canada and the U.S., and smokers’ actual intake of these toxins. (7),(8),(9),(10)

These researchers have noted that the machines used to test the cigarettes do not adequately mimic real smokers’ behaviour and, in many cases, significantly underrepresent smokers’ actual intake of toxins. Because reported levels do not represent absolute quantities of substances in the cigarette or cigarette smoke, intake is reliant upon smoking behaviour. Puff duration, frequency, and quantity will vary from smoker to smoker, resulting in varying yields.

Epidemiologic data appear to confirm that the development and increasing popularity of “low tar/nicotine” cigarettes has not led to lower risk of cigarette- caused diseases.(11)

But what if reported tar, nicotine, and carbon monoxide levels are taken at face value? How are descriptive names used by manufacturers to convey differences in reported tar, nicotine, and carbon monoxide levels to consumers? Even if reported levels were what smokers actually ingested, are smokers still being “duped” as well as “doped?”

Method

The Tobacco Products Control Act and Regulations require manufacturers and importers of cigarettes, cigarette tobacco and smokeless tobacco to report to the government tar, nicotine and carbon monoxide levels in their products as measured by defined methods.

These data were obtained from Health Canada from various years. Data from 1995, the most recent year summarized, were reviewed for this paper. Levels of all reported toxins — tar, nicotine and carbon monoxide — were examined within descriptor categories (regular size “light”, king size “special mild”, etc.) and within brand families with four or more brands.

Categories were assigned an arbitrary classification based on percentage variation of levels of tar, nicotine and carbon monoxide within the category: less than 25% variation was classified as “low,” 25-39% as “moderate,”40-54% as “high,” and 55% or greater as “very high.”

Results

Variation Within Descriptor Categories

Most descriptor categories contain a wide range of reported levels of toxins. In many cases, the range of toxic constituent levels was greater within a descriptor category than within large brand families. This is summarized in Table 2 and Figure 1.

A high degree of variation was found regardless of the number of brands in the category. The widest variation was found in the king size filter tip “ultra mild” category, with tar levels varying by 77%, nicotine by 70%, and carbon monoxide by 83%. The most consistent category was regular size filter tip “medium”, with variations of 7%, 8%, and 7% respectively for tar, nicotine, and carbon monoxide.

Variation Within Brand Families

The level of variation within brand families was, in general, quite high. Most brand families — du Maurier, Matinée, Export “A”, Macdonald, Vantage, Craven “A”, Rothmans and Viscount — contained brands with high or very high variation in reported levels of toxins. Brand families with low to moderate variation were Player’s (see Figure 2), Belvedere, Mark Ten, Number 7 and Sportsman.

Within brand families, two predictors of levels of toxins ought logically to be size (regular vs. king size) and descriptor (light vs. extra light). However, inconsistencies were common.

For example, du Maurier regular size filter tip “light” cigarettes yield more reported carbon monoxide than the king size equivalent. Player’s king size filter tip “medium”, regular size filter tip “medium”, and king size filter tip “light” cigarettes all yield identical reported nicotine levels. Export “A” regular size filter tip “light” cigarettes yield higher levels of all toxins than the king size equivalent. The same is true for Mark Ten regular and king size filter tip “lights”.

In many cases where the levels of some toxins varied consistent with the descriptor, the differences were negligible, or non-existent for other toxins. For example, Export “A” regular size filter tip “medium” and “light” cigarettes have tar levels that vary by one mg, but identical nicotine and carbon monoxide levels.

Discussion and Conclusions

Descriptive language on cigarette labels at best confuses and at worst misleads. Reported levels of toxins within descriptor categories vary widely in most cases. Tobacco company officials have testified that descriptors are not meant to encompass similar reported tar and nicotine levels, but rather to indicate relative levels within brand families.(12) However, the data reveals that descriptors are not necessarily consistent or informative even in this context.

In addition, the commonly moderate variations in toxins within brand families (between “light”, “extra light” and “ultra light”, for example) are unlikely to translate into different levels when actually smoked, given documented variation in actual smoking patterns.

The labelling of tobacco products is treated differently from any other ingested consumer product. Manufacturers of both food and alcohol products must adhere to federally regulated guidelines when using descriptive language on product labels. Tobacco products are not under similar specific obligations. They are, however, subject to the general provisions of the Consumer Products and Labelling Act, which specifically prohibits labels “containing any false or misleading representation that relates to or may reasonably be regarded as relating to that product.” (s. 7(1)) (13)

There is sufficient evidence to suggest that a review of tobacco product labelling in relation to this Act is in order. Yet when complaints have been officially lodged,(14) the government has failed to restrict the way in which tobacco manufacturers use descriptive labelling.

Of greater importance to public health, however, are the measurement systems in place from which the reported toxic yields are derived and how consumers perceive both the levels and the descriptive labelling. The reported levels in and of themselves are misleading because they do not necessarily represent what all, or even most, smokers ingest. The manufacturers take advantage of the flawed system to label and market their products in order to appeal to as many smokers as possible.

It is clear that, at the very least, descriptive labelling on tobacco packaging requires regulation. More importantly, the way in which tar, nicotine and carbon monoxide levels are measured should be re-evaluated and changed to more accurately reflect the levels ingested by smokers. Smokers have a right to this information, but it must be presented in a manner which does not lead to misperceptions about relative health and safety risks.

Notes

1. The Creative Research Group Limited for Imperial Tobacco Limited. Project Viking, Volume II: An Attitudinal Model of Smoking (Exhibit AG-21B in RJR MacDonald Inc. vs. Attorney General of Canada), 1986, 1.
2. Imperial Tobacco Limited. Response of the Market and of Imperial Tobacco to the Smoking and Health Environment (Exhibit AG-41 in RJR MacDonald Inc. vs. Attorney General of Canada), 1978, 2.
3. National Cancer Institute. The FTC Cigarette Test Method for Determining Tar, Nicotine, and Carbon Monoxide Yields of U.S. Cigarettes. Report of the NCI Expert Committee.Smoking and Tobacco Control Monograph #7. United States: National Cancer Institute, National Institutes of Health, Public Health Service, U.S. Department of Health and Human Services, August 1996.
4. Imperial Tobacco Limited. Fiscal ‘81 Media Plans (Exhibit AG-223 in RJR MacDonald Inc. vs. Attorney General of Canada), 1979.
5. RJR-MacDonald Inc. vs. Attorney General of Canada, Vol. 4, 27 September 1989, 507.
6. Gallup Organization, Inc. The Public’s Attitudes Toward Cigarette Advertising and Cigarette Tax Increase. Princeton, NJ: The Gallup Organization, Inc., 1993.
7. National Cancer Institute. The FTC Cigarette Test Method for Determining Tar, Nicotine, and Carbon Monoxide Yields of U.S. Cigarettes. Report of the NCI Expert Committee.Smoking and Tobacco Control Monograph #7. United States: National Cancer Institute, National Institutes of Health, Public Health Service, U.S. Department of Health and Human Services, August 1996.
8. Kozlowski, LT, Pope, MA, and Lux, JE. “Prevalence of the Misuse of Ultra- Low-Tar Cigarettes by Blocking Filter Vents.” American Journal of Public Health, 78(6), 694-695, 1988.
9. Kozlowski, LT. “Tar and Nicotine Delivery of Cigarettes.” Journal of the American Medical Association, 245(2), 158-159, 1981.
10. Djordjevic, MV, Fan, J, Ferguson, S, and Hoffmann, D. “Self-regulation of smoking intensity. Smoke yields of the low-nicotine, low-’tar’ cigarettes.” Carcinogenesis, 16(9),2015-2021, 1995.
11. U.S. Department of Health and Human Services. Reducing the Health Consequences of Smoking: 25 Years of Progress. A Report of the Surgeon General.. U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control, Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health. DHHS, 1989.
12. RJR-MacDonald Inc. vs. Attorney General of Canada, Vol. 8, 4 October 1989, 1156.
13. R.S.C. 1985, c. C-38.
14. The Non-Smokers’ Rights Association filed a complaint in December 1990 with Consumer and Corporate Affairs Canada (CCAC) claiming that certain tobacco product labelling was in violation of the Consumer Packaging and Labelling Act. In February 1992, CCAC concluded its investigation, deciding not to pursue enforcement action. No explanation of the decision was provided.

Table 1

Descriptors Used in Cigarette Labelling in Canada, 1995

Table 2

Range and Variation of Toxic Yields Within Descriptor Category

(in descending order)

???

Figure 1

Figure 2

Table 4

Range and Variation of Toxic Yields Within Brand Family

(in descending order)

Table 5

Intra-brand Family Comparisons

The smaller size cigarette yields higher reported toxins.

Appendix A

1995 Data – By Descriptor