Cigarette Smoking Among Adults-United States, 1998

One of the national health objectives for 2010 is to reducethe prevalence of cigarette smoking among adults to no more than 12% (objective21.la). To assess progress toward meeting this objective, the Centers forDisease Control and Prevention (CDC) analyzed self-reported data from the 1998National Health Interview Survey (NHIS) Sample Adult Core Questionnaire aboutcigarette smoking among adults in the United States. This report summarizes thefindings of this analysis, which indicate that in 1998, 24.1% of adults werecurrent smokers.

The 1998 NHIS Core Questionnaire was administered to anationally representative sample (n = 32,440) ofthe US noninstitutionalized civilian population aged ³ 18 years; theoverall response rate for the survey was 73.9%. Participants were asked, "Haveyou smoked at least 100 cigarettesin your entire life?" and "Do younow smoke cigarettes every day,some days, or not at all?" Current smokers were persons who reported bothhaving smoked ³ 100 cigarettes during their lifetime and having smoked everyday or some days at the time of the interview. Former smokers were those who hadsmoked ³ 100 cigarettes during their lifetime but did not currently smoke.

Attempts to quit were determined by asking current smokers,"During the past 12 months, have you stopped smoking for one day or longerbecause you were trying to stop smoking?" Data were adjusted fornonresponse and weighted to provide national estimates. Confidence intervals(CIs) were calculated using SUDAAN software for statistical analysis.

Prevalence Lowest AmongOlder and Educated Adults

In 1998, an estimated 47.2 million adults (24.1%), comprising24.8 million men (26.4%) and 22.4 million women (22.0%), were current smokers (Table 1). Overall, 19.7% (95% CI =± 0.6) of adults were "every-day smokers," and 4.2% (95% CI = ± 0.3)were "some-day smokers" (every-day smokers constituted 82.4% [95% CI =± 1.01] of all smokers).

Prevalence of smoking was highest among persons aged 18 to 24years (27.9%) and aged 25 to 44 years (27.5%), and lowest among persons aged ³65 years (10.9%). Prevalence of current smoking was highest among AmericanIndians/Alaska Natives (40.0%), intermediate among non-Hispanic whites (25.0%)and non-Hispanic blacks (24.7%), and lowest among Hispanics (19.1%) andAsians/Pacific Islanders (13.7%).

Adults with ³ 16 years of education had the lowest smokingprevalence (11.3%), achieving the 2010 goal of reducing smoking rates to no morethan 12%. Current smoking prevalence was highest among persons with 9 to 11years of education (36.8%). Smoking prevalence was higher among persons livingbelow the poverty level (1997 poverty thresholds from the Bureau of the Census,Economics, and Statistics Administration, US Department of Commerce, were usedin these calculations.) (32.3%) than among those living at or above the povertylevel (23.5%).

In 1998, an estimated 44.8 million adults (22.9% [95% CI = ±0.6]) were former smokers, comprising 25.7 million men and 19.1 million women.Former smokers constituted 48.7% (95% CI = ± 1.0) of persons who had eversmoked ³ 100 cigarettes. Among current daily smokers in 1998, an estimated 15.2million (39.2% [95% CI = ± 1.4]) had stopped smoking for at least 1 day duringthe preceding 12 months because they were trying to stop smoking.

Editorial Note From the CDC

The findings in this report suggest that the goal of reducingthe prevalence of cigarette smoking among adults to ≤ 12% by 2010 will requireaggressive public health efforts to implement comprehensive tobacco-controlprograms nationwide. The 1998 NHIS data also demonstrate substantial differencesin smoking prevalence across populations.

In 1998, smoking prevalence among persons aged 18 to 24 yearswas as high as the prevalence among persons aged 25 to 44 years. Historically,smoking prevalence has been highest among persons aged 25 to 44 years andsignificantly lower among persons aged 18 to 24 years. Recent increases amongpersons aged 18 to 24 years may reflect the aging of the cohort of high-schoolstudents among whom current smoking rates were high during the 1990s. Inaddition, the increase may indicate increased initiation of smoking among youngadults. The high prevalence of smoking among young adults indicates a need tofocus tobacco-use prevention and treatment programs on both adolescents andyoung adults.

Smoking Prevalence Highest Among Native Americans

Smoking prevalence reported for racial/ethnic subgroups showedfew changes from 1997 to 1998. Prevalence of current smoking among AmericanIndians/Alaska Natives remained the highest in the nation. State and regionalsurveys indicate that the prevalence of smoking cessation among AmericanIndians/Alaska Natives remains relatively low. Although many factors contributeto the high prevalence of smoking among American Indians/Alaska Natives, it isimportant to develop culturally appropriate prevention and control measures thatdistinguish between the use of manufactured tobacco products and the ceremonialuse of tobacco.

National health objectives for 2010 that are focused oneliminating population disparities reinforce the need for greater surveillanceand culturally responsive approaches to tobacco use across communities. In theUnited States, population disparities in smoking prevalence have been consistentfrom 1993 through 1998. For example, in 1993, an 8.3 (95% CI = ± 2.5)percentage-point difference in smoking prevalence existed between those at orabove the poverty level and those below (23.8% and 32.1%, respectively). In1998, the difference was 8.8 (95% CI = ± 1.9) percentage points (23.5% and32.3%, respectively).

Similarly, differences in prevalence among various educationalgroups have not been reduced. In 1993, the difference between those with 9 to 11years of education and those with ³ 16 years was 23.3 (95% CI= ± 3.0)percentage points (36.8% and 13.5%, respectively). In 1998, the difference was25.5 (95% CI= ± 2.3) percentage points (36.8% and 11.3%, respectively). Therelation between tobacco use and increased risk of failing or dropping out ofhigh school demonstrates the necessity of reaching these students throughschool-based programs before they leave school.

Differences in prevalence among racial/ethnic subgroups have notbeen reduced. For example, in 1993, the difference between non-Hispanic whitesand American lndians/Alaska Natives was 13.3 (95% CI = ± 8.7) percentage points(25.4% and 38.7%, respectively). In 1998, the difference between non-Hispanicwhites and American Indians/Alaska Natives was 15.0 (95% CI= ± 9.8)percentage points (25.0% and 40.0%, respectively). The reduction oftobacco-related health disparities requires communities, states, and nationalorganizations to take a multidisciplinary approach to tobacco prevention andcontrol.

Limitations of Findings

The findings in this report are subject to at least twolimitations. Because the questionnaire for the 1997 NHIS was redesignedcompletely, trend analysis or comparison with data from years before 1997 shouldbe conducted with caution. Second, the sample size of certain subgroups (forexample, American Indians/Alaska Natives) was small, possibly resulting inunstable estimates.

Although comprehensive programs are critical in reducing theburden of tobacco use, short-term decreases in tobacco-related morbidity andmortality can be achieved only by helping current smokers quit. To assist inthis process, the US Department of Health and Human Services has releasedguidelines with specific evidence-based recommendations for tobacco-usetreatment. Recommended interventions include individual, group, or telephonecounseling that offers practical advice about and support for quitting; supportfrom family and friends also improves success rates.

In addition, all smokers trying to quit should be encouraged touse a medication approved by the Food and Drug Administration—either nicotinereplacement therapy (gum, inhaler, nasal spray, or patch) or a nonnicotinepharmacologic aid (such as bupropion [Zyban]). To ensure that smokers interestedin quitting receive appropriate treatment, health-care systems must make routinescreening of tobacco use the standard of care and monitor (through qualityassurance processes) the provision of appropriate interventions to smokers.Improving access to smoking cessation treatment by reducing cost barriers alsoincreases the number of quitters.

A comprehensive approach to tobacco control will requiretreatment for nicotine dependence and efforts at national, state, and locallevels to reduce youth smoking, promote smoke-free environments, supportcounter-marketing efforts, enforce laws and regulations, and eliminate disparities intobacco use among population subgroups. Increased attention must be focused ongroups that show no decline in smoking prevalence, including persons aged 18 to24 years, adults with low education levels, and American Indians/Alaska Natives.Approaches with the widest scope (ie, economic, regulatory, and comprehensive)are likely to have the greatest long-term population impact.