Tobacco Use Among Middle and High School Students
-- United States, 1999

Citizens for Clean Air and Clean Lungs

Morbidity and Mortality Weekly Report
January 28, 2000 / 49(03); p.49-53

The prevalence of cigarette smoking nationwide among high school students increased during the 1990s; more than 80% of current adult tobacco users started smoking cigarettes before age 18 years. To determine the prevalence of cigarette, smokeless tobacco (i.e., chewing tobacco and snuff), cigar, pipe, bidi, and kretek use among middle school and high school students nationwide, the American Legacy Foundation, in collaboration with the CDC Foundation, conducted the National Youth Tobacco Survey (NYTS) during the fall of 1999.

smoking warning

This report summarizes data from the NYTS on current use of tobacco products, which indicate that 12.8% of middle school students and 34.8% of high school students use any type of tobacco; that the low prevalence of current cigarette smoking observed among black high school students throughout the 1990s is not found among middle school students; and that the percentages of high school students who currently use bidis and kreteks (two new forms of tobacco in the United States) are almost as high as the proportion who use smokeless tobacco.

The school-based 1999 NYTS employed a nationally representative sample of students in grades 6-12. 145 schools were selected, and approximately five intact classes of a required subject (e.g., English or social studies) across grades 6-12 were randomly selected from each participating school. All students in the selected classes were eligible to participate. 15,058 students in 131 schools completed questionnaires. The school response rate was 90%, and the student response rate was 93%, resulting in an overall response rate of 84%.

Students completed an anonymous, self-administered questionnaire that included questions about tobacco use, exposure to environmental tobacco smoke, minors' ability to purchase or otherwise obtain tobacco products, knowledge and attitudes about tobacco, and familiarity with pro- and anti-tobacco media messages. Current use of bidis, cigarettes, cigars, kreteks, pipes, and smokeless tobacco was defined as use on one or more of the 30 days preceding the survey. Any current tobacco use was defined as using any of these products on one or more of the 30 days preceding the survey.

Middle School Students
Among middle school (grades 6-8) students, the overall prevalence of any current tobacco use was 12.8% (see Table 1 below). Cigarettes (9.2%) were the most prevalent type of tobacco used, followed by cigars (6.1%). Cigarette smoking rates were similar among boys and girls and among racial/ethnic groups.

Boys were significantly more likely than girls to use smokeless tobacco (4.2% and 1.3%, respectively), smoke cigars (7.8% and 4.4%, respectively), and smoke tobacco in a pipe (3.5% and 1.4%, respectively). Black students were significantly more likely than white students to smoke cigars (8.8% and 4.9%, respectively).

High School Students
Among high school (grades 9-12) students, the overall prevalence of any current tobacco use was 34.8%. Cigarettes (28.4%) were the most prevalent type of tobacco used, followed by cigars (15.3%). Boys were significantly more likely than girls to use smokeless tobacco (11.6% and 1.5%, respectively), smoke cigars (20.3% and 10.2%, respectively), smoke tobacco in a pipe (4.2% and 1.4%, respectively), and smoke bidis (6.1% and 3.8%, respectively).

White and Hispanic students were significantly more likely than black students to smoke cigarettes (32.8%, 25.8%, and 15.8%, respectively). White students were significantly more likely than black and Hispanic students to use smokeless tobacco (8.7%, 2.4%, and 3.6%, respectively).


Percentage of students in middle school (grades 6-8) and high school (grades 9-12)
currently* using tobacco products, by type of tobacco product, sex, and race/ethnicity
United States, National Youth Tobacco Survey, 1999
[95% Confidence intervals in parentheses. Middle School = MS; High School = HS]

* Used tobacco on one or more of the 30 days preceding the survey

A. Any Tobacco Use (cigarettes, smokeless tobacco, cigars, pipes, bidis, or kreteks)
Sex Race/Ethnicity
  Male Female White Black Hispanic Total
MS 14.2 ( 2.2) 11.3 ( 2.2) 11.6 ( 2.3) 14.4 ( 2.7) 15.2 ( 5.2) 12.8 ( 2.0)
HS 38.1 ( 3.2) 31.4 ( 3.1) 39.4 ( 3.2) 24.0 ( 4.2) 30.7 ( 4.4) 34.8 ( 2.7)
B. Cigarettes
  Male Female White Black Hispanic Total
MS 9.6 ( 1.7) 8.8 ( 1.7) 8.8 ( 2.0) 9.0 ( 1.8) 11.0 ( 4.1) 9.2 ( 1.6)
HS 28.7 ( 2.8) 28.2 ( 3.3) 32.8 ( 3.1) 15.8 ( 3.8) 25.8 ( 4.7) 28.4 ( 2.7)
C. Smokeless Tobacco/Chew/Snuff
  Male Female White Black Hispanic Total
MS 4.2 ( 1.3) 1.3 ( 0.5) 3.0 ( 1.1) 1.9 ( 0.9) 2.2 ( 0.9) 2.7 ( 0.7)
HS 11.6 ( 2.8) 1.5 ( 0.6) 8.7 ( 2.1) 2.4 ( 1.3) 3.6 ( 1.6) 6.6 ( 1.6)
D. Cigars
  Male Female White Black Hispanic Total
MS 7.8 ( 1.3) 4.4 ( 1.3) 4.9 ( 1.0) 8.8 ( 2.3) 7.6 ( 2.9) 6.1 ( 1.1)
HS 20.3 ( 1.9) 10.2 ( 1.6) 16.0 ( 1.6) 14.8 ( 3.5) 13.4 ( 2.9) 15.3 ( 1.4)
E. Pipe
  Male Female White Black Hispanic Total
MS 3.5 ( 0.8) 1.4 ( 0.6) 2.0 ( 0.6) 2.0 ( 0.9) 3.8 ( 1.7) 2.4 ( 0.5)
HS 4.2 ( 0.9) 1.4 ( 0.5) 2.6 ( 0.6) 1.8 ( 0.9) 3.8 ( 1.4) 2.8 ( 0.5)
F. Bidis
  Male Female White Black Hispanic Total
MS 3.1 ( 0.8) 1.8 ( 0.6) 1.8 ( 0.5) 2.8 ( 1.3) 3.5 ( 1.6) 2.4 ( 0.6)
HS 6.1 ( 1.0) 3.8 ( 1.0) 4.4 ( 0.9) 5.8 ( 2.1) 5.6 ( 2.1) 5.0 ( 0.8)
G. Kreteks
  Male Female White Black Hispanic Total
MS 2.2 ( 0.6) 1.7 ( 0.7) 1.7 ( 0.7) 1.7 ( 0.8) 2.1 ( 0.6) 1.9 ( 0.5)
HS 6.2 ( 1.1) 5.3 ( 1.5) 6.5 ( 1.5) 2.8 ( 1.5) 5.5 ( 1.9) 5.8 ( 1.2)

CDC Editorial Note
This report is the first to measure the prevalence of current tobacco use among a nationally representative sample of middle school students and the first to report the prevalence of current bidi and kretek use among a nationally representative sample of middle and high school students.

Although previous national surveys have shown that cigarette smoking rates among black high school students have been increasing, black students still were smoking at much lower rates than other high school students. However, the findings in this report indicate that current cigarette smoking prevalence among middle school black students was similar to rates among white and Hispanic students and that current cigar use prevalence among middle school black students was significantly higher than among white students.

Future surveys should evaluate whether the rate of increase in smoking rates among black students has accelerated and whether the difference in smoking rates between black and white high school students are disappearing. In addition, more research is needed to determine whether black youth are finding smoking appealing and socially acceptable.

Current use of novel tobacco products, such as bidis and kreteks, is an emerging public health problem among U.S. youth. Cigarettes remain the most widely used tobacco product by youth; however, recent trends underscore the importance of monitoring the rates at which youth adopt other tobacco products. The social and cultural factors related to differing patterns of tobacco product use across sex and racial/ethnic groups require additional study.

The 1999 NYTS estimates for high school students will be compared with those of the Monitoring The Future (MTF) study and the Youth Risk Behavior Survey (YRBS), the other national school-based surveys. Comparison of NYTS estimates with those of other national surveys must be interpreted with caution for several reasons.

First, YRBS and MTF were conducted during spring 1999, and NYTS was conducted during September-October 1999, a different academic year. Within each grade, the fall school population is approximately 6 months younger than the spring school population. This difference can be expected to lead to higher estimates of ever smoking in the spring surveys and may lead to higher estimates of current smoking.

Second, the tobacco industry increased the wholesale price of tobacco products during 1999, but also provided substantial price discounts during the same period, making determination of the precise effect of retail prices on smoking rates difficult. However, preliminary per capita consumption estimates from the U.S. Department of Agriculture suggest cigarette consumption has decreased in 1999, suggesting that the prevalence among youth also may have decreased.

Third, the NYTS is a single-topic survey (tobacco), and MTF and YRBS are multi-topic surveys. The effect of the number of topics surveyed on the resulting estimates is unknown. Finally, NYTS had a 90% school response rate, a higher reported school response rate than YRBS and MTF. Some schools that participated in the NYTS may not participate in YRBS or MTF.

The findings in this report are subject to at least two limitations. First, these data apply only to youth who attended middle or high school and are not representative of all persons in this age group. Few persons aged less than 16 years do not attend school and, in 1997, only 4% of 16-year-olds and 6% of 17-year-olds who had not completed high school were not enrolled in a high school program. The dropout rate for young adults aged 16-24 years varies greatly by race/ethnicity (7.6%, white; 13.4%, black; and 25.3%, Hispanic). Second, "any current tobacco use" might be underestimated in this report because it does not include a measure of "roll-your-own" tobacco smoking.

To evaluate the potential impact of the expanding levels of tobacco prevention efforts nationwide and in the individual states, surveillance of trends in tobacco use among youth must be continued and expanded. YRBS has provided national and state-specific surveillance of tobacco use among high school students since 1991. The NYTS and state-specific youth tobacco surveys are extending this surveillance effort to middle school students and across a wider range of evaluation variables, including knowledge and attitudes about tobacco, exposure to environmental tobacco smoke, familiarity with pro-smoking and antismoking media messages, and exposure to tobacco-use prevention curriculum in schools.

CDC has prepared "Best Practices" guidelines to help states determine funding priorities and to plan and carry out effective comprehensive tobacco-use prevention and control programs. If current patterns of smoking behavior persist, an estimated 5 million U.S. persons who were aged less than or equal to 18 years in 1995 could die prematurely from smoking-related illnesses.

Implementation of the "Best Practices" guidelines, along with nationwide prevention efforts, enforcement of the proposed Food and Drug Administration rules, increases in the excise tax on tobacco products, and increased availability of smoking cessation treatment options, could dramatically reduce these projected deaths.

Click here for the full text of this article.

Editorial Note
The 1999 National Youth Tobacco Survey (NYTS) is the first national school-based survey to provide information not only about tobacco use, but about related knowledge, beliefs, attitudes and behaviors. It is also the first national survey to focus on all varieties of tobacco use among youth and the first survey of middle school students.

As such, this survey will provide us with considerable new knowledge about the onset of the nicotine epidemic and should serve to remind us not to forget about all forms of tobacco use. We look forward to reading future reports on the 1999 survey and hope that future surveys will be conducted to enable us to examine trends over time.

The prevalence of current cigarette use among high school students in Fall 1999 has decreased substantially since Spring 1997. These comparisons are shown in Table 2 below.

  1997 YRBS 1999 NYTS
Overall 36.4% ( 2.3) 28.4% ( 2.7)
Males 37.7% ( 2.7) 28.7% ( 2.8)
Females 34.7% ( 2.8) 28.2% ( 3.3)
Whites 39.7% ( 2.4) 32.8% ( 3.1)
Blacks 22.7% ( 3.8) 15.8% ( 3.8)
Hispanics 34.0% ( 2.7) 25.8% ( 4.7)

Current smokeless tobacco among male high school students declined from 15.8% ( 3.7) in 1997 to 11.6% ( 2.8) in 1999; however, this decline was not statistically significant (as the 95% confidence intervals overlap). The declines in cigarette use may be due to events that occurred in the interim (e.g. price increases due to the Tobacco Settlement or state tax increases) or to cohort differences. However, it's also possible that the differences in smoking prevalence are at least partially attributable to differences in the methods used in these two surveys (e.g., different content of survey questionnaires and response rates).

The National Youth Tobacco Survey (NYTS) appears to have employed very high quality survey methods. Sampling, survey administration, response rates, and weighting and analysis of the data appear to be very strong. The overall response rate of 84% boosts our confidence that the results from this sample are generalizable to the national school population. (The report, however, did not indicate whether private schools were included in the sample.) The overall response rate and the school participation rate (90%) for this survey are remarkable especially as compared to other national school-based surveys (e.g., the overall response rate for the 1997 National Youth Risk Behavior Survey (YRBS) was 69% and for the 1998 Monitoring the Future Survey (MTF) was less than 58%). Also remarkable is how quickly the preliminary findings from this survey were released to the public. Perhaps, other national surveys can be improved in these areas.

The 95% confidence interval means that there is a 95% likelihood that the actual value in the population should be within that many percentage points of the observed sample value. For example, 12.8 ( 2.0) for total tobacco use among middle school students (which appears at the end of the first row of Table 1) can be interpreted as follows: there is a 95% likelihood that from 10.8% (12.8 - 2.0) to 14.8% (12.8 + 2.0) of middle school students in the U.S. used some form of tobacco in the 30 days preceding the survey.