The study found that extreme dieting was an independent predictor of smoking. Extreme dieting involved going without eating for 24 hours or more in the past 30 days in order to lose weight or keep from gaining it, the taking of diet aids during the past 30 days, and inducing vomiting or taking laxatives to lose weight. Extreme dieters showed a higher variability in smoking behavior than their peers. The trend analysis showed that the magnitude of the association between smoking and extreme dieting was smaller in recent years among adolescents, but remained unchanged among non-overweight girls over the same period. These findings suggest that when adolescent smoking behavior is examined, the intensity of dieting behavior should be considered, along with other co-occurring unhealthy behaviors. The study analyzed the Youth Risk Behavior Survey datasets, using the multivariable logistic regression method. The sample size of each year of the survey ranged from 13,554 to 15,273, with girls composing 49-51 percent of the total sample (71,854). Approximately 62 percent of the study participants were White, and 14 percent were Black. Prevalence estimates of current smoking and corresponding 95-percent confidence intervals were computed across four comparison groups based on gender and body weight. 3 tables, 1 figure, and 38 references