The evaluation of efforts to prevent tobacco use among adolescents
requires accurate surveillance of both smoking prevalence and
smoking initiation rates. Although several surveillance systems
provide timely data about adolescent smoking prevalence, data
characterizing rates of smoking initiation among adolescents have
been limited. To improve characterization of trends in smoking
initiation among young persons, data from the Tobacco Use Supplement
of the 1992 and 1993 Current Population Surveys (CPS) were used
to estimate smoking initiation rates for persons who were adolescents
(age 14 to 17 years) or young adults (age 18 to 21 years) during
1980 to 1989. This report summarizes the results of that analysis.
The CPS are monthly surveys of the US civilian, noninstitutionalized
population age equal to or more than 15 years. Approximately 56,000
households are surveyed each month; one household respondent provides
information about all household members age equal to or more than
15 years. Questions about tobacco use were added to the September
1992, January 1993, and May 1993 monthly surveys. The response
rates for the three surveys were 84.7%, 84.9%, and 82.0%, respectively
(N = 293,543 household members). To minimize biases that could
result from discrepancies between self-reports and proxy reports
of smoking behavior, this analysis used data from self-respondents
only (82% of total sample).
Ever-smokers were defined as respondents who answered "yes"
to the question, "Have you smoked at least 100 cigarettes
in your entire life?" Ever-smokers were asked, "How
old were you when you started smoking cigarettes fairly regularly?"
To restrict the analysis to persons who were adolescents or young
adults for some period during 1980-1989, only respondents aged
17-34 years at interview were included. The final sample consisted
of 71,321 persons, of whom 27,768 (38.9%) were ever-smokers.
Using the age of respondents at the time of the interview and
the age they reported starting smoking, the age of respondents
and their smoking status were calculated for each year during
the 1980s. The denominator for the initiation rate for a given
year was the number of respondents at risk for initiating smoking
during that year (persons already smoking were eliminated from
the denominator for that year). The numerator was the number of
respondents who reported initiating smoking during that year.
Data were weighted by age, sex, and race/ethnicity to provide
Among adolescents, the smoking initiation rate decreased slightly
from 1980 (5.4%) through 1984 (4.7%) and then increased through
1989 (5.5%); the largest annual increase occurred in 1988 (Figure
1). In comparison, among young adults, initiation rates decreased
throughout the 1980s (Figure 1). For both age groups, initiation
rates and trends were similar for males and females.
Editorial Note From the CDC
The findings in this report indicate an increase in the rate of
initiation of cigarette smoking among adolescents from 1985 through
1989, a period during which the rate among young adults declined
and overall prevalence of smoking among adults decreased steadily.
One important consequence of the increased rate of initiation
among adolescents will be the increased future burden of tobacco-related
disease. In particular, because of the increase in initiation
since 1984, it is estimated that an additional 600,000 adolescents
began to smoke during 1985 to 1989. Of those adolescents who continue
to smoke regularly, approximately 50% will die from smoking-attributable
Potential reasons for an increase in smoking initiation rates
among adolescents include a decreased real price of cigarettes,
increased levels of disposable income, increased acceptability
of smoking, and intensified cigarette marketing. However, because
the real price of cigarettes increased steadily during 1985 to
1989 and the real average weekly income among high school seniors
remained stable during this period, cigarettes were less affordable
to young persons. In addition, the acceptability of smoking among
high school seniors did not increase: During this period there
were increases in the percentages of high school seniors who believed
that cigarettes are harmful, smoking is a "dirty habit,"
and becoming a smoker reflects poor judgment, and who reported
that they "mind being around people who are smoking"
and would prefer to date nonsmokers.
Is Advertising the Cause?
The increase in rates of smoking initiation among adolescents
during 1985 to 1989 may reflect increased real expenditures for
cigarette advertising and promotion. The increase in rates occurred
during a period when real expenditures for total cigarette advertising
and promotion* doubled, and expenditures for cigarette
promotion more than quadrupled (Figure 2): from 1980 to 1989,
total annual advertising and promotional expenditures (in 1993
dollars) increased from $2.1 billion to $4.2 billion, while promotional
expenditures alone increased from $771 million (37% of total expenditures)
to $3.2 billion (76%) (Figure 2). Promotional efforts have been
highly effective among adolescents. For exam ple, among persons
age 12 to 17 years in 1992, approximately 50% of smokers and 25%
of nonsmokers reported having received promotional items from
An association between overall cigarette marketing expenditures
and initiation rates for smoking among adolescents is plausible
for at least four reasons:
- First, brand loyalty is usually established with the first
cigarette smoked; therefore, cigarette companies have an economic
incentive to encourage first-time smokers to smoke their brands.
- Second, adolescents are exposed to cigarette advertising and
promotions that employ themes and images that appeal to young
- Third, advertising directly influences brand awareness and
attitudes toward smoking among adolescents. Specifically, adolescents
smoke the most heavily advertised brands, and changes in brand
preferences among young persons are associated with changes in
brand-specific advertising expenditures. For example, the Joe
Camel campaign introduced nationally in 1988 was associated with
an increase in the market share of that specific brand among adolescents.
- Finally, consumer research suggests that younger persons (ie,
age 14 to 17 years) aspire to be young adults; therefore, advertising
and promotional efforts targeted toward young adults may have
greater appeal to adolescents because of their age aspirations.
Although current estimates of smoking initiation rates among adolescents
are not available, from 1991 through 1993, the national prevalence
of smoking increased among eighth and tenth-grade students. To
reverse the trend of increasing smoking initiation rates among
adolescents and to achieve the national health objective for the
year 2000 of reducing the initiation of cigarette smoking by youth
(no more than 15% should become regular smokers by age 20; objective
3.5), prevention efforts that focus on young persons should be
intensified. Such efforts could include making cigarettes less
affordable by either increasing their real price or by limiting
sales to cartons rather than individual packs, enforcing laws
prohibiting the sale and distribution of cigarettes to young persons,
conducting mass media campaigns to discourage tobacco use, and
eliminating or severely restricting all forms of tobacco product
advertising and promotion to which young persons are likely to
Reported by: K.M. Cummings, PhD, D Shah, MS, Roswell Park
Cancer Institute, Buffalo, New York; D.R. Shopland, National Cancer
Institutes of Health, Office on Smoking and Health, National Center
for Cronic Disease Prevention and Health Promotion, CDC.
Adapted from Morbidity and Mortality Weekly Report,vol 44,