Smoking and Tobacco Control Monograph No. 7 
market share, suggesting an attempt to increase market share. As shown in 
Figure 2, the two proportions are converging. In 1992 lower tar cigarettes 
accounted for 69 percent of market share and 71 percent of advertising 
expenditures (Federal Trade Commission, 1994). 
One major purpose of the marketing of these varieties of cigarettes 
appears to have been to alleviate smokers' health concerns (Pollay, 1990; 
Warner 1985). The advertisements seem to have achieved a large part of 
their goal. In 1993 a Gallup Organization poll posed the following question: 
"Besides selling the product, what message do you think cigarette advertising 
is trying to get across when it uses terms like low tar, low nicotine, or low 
yield?" (Gallup Organization, Inc., 1993, pp. 22). Fifty-eight percent of 
respondents (56 percent of smokers and 60 percent of nonsmokers) answered 
that the message indicates a positive health benefit, that is, that the brand 
is safer, healthier, less harmful, not as bad for you, or less cancerous (Gallup 
Organization, Inc., 1993). 
MONITORING Three national surveys helped shed light on the patterns in j 
NATIONAL DATA attitudes, knowledge, and beliefs about low-yield cigarettes: i 
the 1986 Adult Use of Tobacco Survey (AUTS), the 1987 National Health 
Interview Survey (NHIS) Cancer Control Supplement, and the 1993 Teenage 
Attitudes and Practices Survey (TAPS). The 1986 AUTS was a national 
telephone survey of approximately 13,000 Americans ages 17 years and ' 
older (Pierce et al., 1990). The nationally representative sample of the 1987 f 
NHIS included about 22,000 Americans ages 18 years and older who were j 
interviewed primarily in their homes (Schoenborn and Boyd, 1989). The I 
1993 TAPS sample included about 13,000 people 10 to 22 years of age ! 
who were contacted via telephone or in their homes (Centers for Disease j 
Control and Prevention, 1994b). The 1993 TAPS included a cross-sectional | 
component of persons 10 to 15 years of age in 1993 and a followup j 
component of a cohort of persons first interviewed in 1989 who were 15 ! 
to 22 years old in 1993 (Centers for Disease Control and Prevention, 1994b). 
There are difficulties in using the 1993 TAPS data to make prevalence | 
estimates. Some participants lost to followup were more likely to be 
smokers in 1989, a phenomenon that would be likely to decrease the overall !| 
prevalence estimate (Centers for Disease Control and Prevention, 1994b). j 
The data used for this report are not used to generate smoking prevalence 
estimates; rather, they look at characteristics of persons who reported that ' 
they were currently smoking. !i 
I 
'Fhe 1986 AU'FS and the 1987 NHIS questions used to determine tar : 
levels assessed items such as brand name, filter vs. nonfilter, pack hardness, f 
cigarette length, mentholation, and if the cigarette was regular, light, or 
ultralight. Fhe tar level assigned is based on responses to the questions using 
FTC tables (Federal Trade Commission, 1985). The tar categories used for i 
this report are (1 ) less than or equal to 6 mg, (2) 7 to 15 mg, and (3) 16 mg 
or more. ( The actual cutpoints used here are 6.99 mg and 15.99 mg.) ^ 
S 
42 
