| Summary: | American students' 30-day smoking prevalence has decreased dramatically over the past two decades. The frequency of smoking within the 30-day measure has shifted from heavy smoking (>1/2 pack/day) toward light smoking (<1 to 5 cigarettes/day). 30-day prevalence thus understates the extent of the decline in youth smoking. To capture this shift toward less frequent smoking among the decreasing proportion of students who smoke, I develop a new index: the average number of cigarettes smoked per student per day (ACSD), using data from Monitoring the Future. To calculate ACSD I assign a specific number of cigarettes to each of 7 response options to the question, “How frequently have you smoked cigarettes during the past 30 days?” Response options range from “not at all” (assigned 0 cigarettes) to “two packs or more per day” (assigned 45 cigarettes). I then multiply these estimates by the proportion of students giving each response option. Summing across the 7 categories produces the ACSD for that survey year. I then compare time trends in 30-day prevalence and ACSD. From the mid-1990s to 2016, 30-day smoking prevalence among 12th graders declined 71.3%, while ACSD dropped 83.9% (p < 0.001). The figures were 84.0% and 90.6% (p < 0.001) for 10th graders and 87.4% and 89.0% for 8th graders (p < 0.05). Sensitivity analysis supports the finding that ACSD has decreased more than 30-day prevalence over time for both 10th and 12th grades. ACSD provides a new measure of the decline in youth smoking to complement the traditional measure of 30-day prevalence. Keywords: Cigarette smoking, Youth, Students, Prevalence, Measurement
|
|---|
