Trends in Cigarette Smoking Among High School Students—United States, 1991-1999

Morbidity and Mortality Weekly Report, August 25, 2000

One of the 10 Leading Health indicators that reflect the major health concerns in the United States is cigarette smoking among adolescents [1]. To examine changes in cigarette smoking among high school students in the United States from 1991 to 1999, CDC analyzed data from the national Youth Risk Behavior Survey (YRBS). This report summarizes the results of the analysis and indicates that current smoking among U.S. high school students increased significantly from 27.5% in 1991 to 34.8% in 1999; however, the analysis also suggested that, later in the decade, current smoking may have leveled or possibly begun to decline.

YRBS measures the prevalence of health risk behaviors among adolescents through representative biennial national, state, and local surveys. The 1991, 1993, 1995, 1997, and 1999 national surveys used independent, three-stage cluster samples to obtain cross-sectional data representative of students in grades 9 through 12 in the 50 states and the District of Columbia. In 1991, 1993, 1995, 1997, and 1999, the respective sample sizes were 12,272, 16,296, 10,904, 16,262, and 15,349; school response rates were 75%, 78%, 70%, 79%, and 77%; student response rates were 90%, 90%, 86%, 87%, and 86%; and overall response rates were 68%, 70%, 60%, 69%, and 66%.

For each cross-sectional survey, students completed an anonymous, self-administered questionnaire that included identically worded questions about cigarette smoking. Lifetime smoking was defined as having ever smoked cigarettes, even one or two puffs. Current smoking was defined as smoking on [greater than or equal to] 1 of the 30 days preceding the survey. Frequent smoking was defined as smoking on [greater than or equal to] 20 of the 30 days preceding the survey. Data are presented only for non-Hispanic black, non-Hispanic white, and Hispanic students because the numbers of students from other racial/ethnic groups were too small for meaningful analysis.

Data were weighted to provide national estimates. SUDAAN was used for all data analysis. Secular trends were analyzed using logistic regression analyses that controlled for sex, race/ethnicity, and grade and that simultaneously assessed linear and quadratic time effects. Quadratic trends suggest a significant but nonlinear trend in the data over time. When a significant quadratic trend accompanies a significant linear trend, the data demonstrate some nonlinear variation (e.g., leveling or change in direction) in addition to a linear trend.

The prevalence of lifetime smoking remained stable from 1991 to 1999 among high school students overall and among all sex, racial/ethnic, and grade subgroups except 10th-grade students. In 1999, 70.4% (95% confidence interval [CI]= [+ or -]3.0) of all students reported lifetime smoking. Among 10th-grade students, lifetime smoking showed a significant linear trend from 1991 (68.3% [95% CI=[+ or -]3.3]) to 1999 (73.9% [95% CI=[+ or -]4.1).

From 1991 to 1999, current smoking exhibited a significant linear trend among students overall and among all sex, racial/ethnic, and grade subgroups (Table 1). The overall prevalence of current smoking was 27.5% in 1991 and 34.8% in 1999. A simultaneous quadratic trend was identified for students overall, suggesting a leveling or possible decline in current smoking. The male, black, black male, and 9th-grade student subgroups also showed this simultaneous quadratic trend.

Each year, white students were significantly more likely than Hispanic students, who were significantly more likely than black students, to report current smoking (except in 1995 when white and Hispanic students were equally likely to report current smoking, but both were significantly more likely than black students to report this behavior). In 1991, white students were 2.5 times more likely than black students and 1.2 times more likely than Hispanic students to report current smoking. In 1999, white students were 2.0 times more likely than black students and 1.2 times more likely than Hispanic students to report current smoking.

The prevalence of frequent smoking showed a significant linear trend from 1991 to 1999 among students overall and in all sex, racial/ethnic, and grade subgroups, except for Hispanic female students. The overall prevalence of frequent smoking was 12.7% (95% CI=[+ or -]2.2) in 1991 and 16.8% (95% CI=[+ or -]2.5) in 1999. Among Hispanic female students, the prevalence of frequent smoking remained stable from 1991 to 1999. For each of the five surveys, white students were significantly more likely than black and Hispanic students to report this behavior.

Reported by: Office on Smoking and Health, and Div of Adolescent and School Health, National Center for Chronic Disease Prevention and Health Promotion, CDC.

Editorial Note: Despite a leveling or possible decline in current smoking among youth overall during the late 1990s, this trend may have been limited to selected groups (i.e., male, black, black male, and 9th-grade students). In addition, frequent smoking rates overall and in all sex, racial/ethnic, and grade subgroups (except Hispanic females) were significantly higher in 1999 than in 1991 and showed no pattern of leveling or declining.