Context: The intensity of smoking, not just prevalence, is associated with future health consequences.
Objective: To estimate smoking intensity patterns over time and by age within birth cohorts for California and the remaining United States.
Design, setting, and participants: Two large population-based surveys with state estimates: National Health Interview Surveys, 1965-1994; and Current Population Survey Tobacco Supplements, 1992-2007. There were 139,176 total respondents for California and 1,662,353 for the remaining United States.
Main outcome measure: Number of cigarettes smoked per day (CPD), high-intensity smokers (≥20 CPD); moderate-intensity smokers (10-19 CPD); low-intensity smokers (0-9 CPD).
Results: In 1965, 23.2% of adults in California (95% confidence interval [CI], 19.6%-26.8%) and 22.9% of adults in the remaining United States (95% CI, 22.1%-23.6%) were high-intensity smokers, representing 56% of all smokers. By 2007, this prevalence was 2.6% (95% CI, 0.0%-5.6%) or 23% of smokers in California and 7.2% (95% CI, 6.4%-8.0%) or 40% of smokers in the remaining United States. Among individuals (US residents excluding California) born between 1920-1929, the prevalence of moderate/high-intensity smoking (≥10 CPD) was 40.5% (95% CI, 38.3%-42.7%) in 1965. Moderate/high-intensity smoking declined across successive birth cohorts, and for the 1970-1979 birth cohort, the highest rate of moderate/high-intensity smoking was 9.7% (95% CI, 7.7%-11.7%) in California and 18.3% (95% CI, 16.4%-20.2%) in the remaining United States. There was a marked decline in moderate/high-intensity smoking at older ages in all cohorts, but this was greater in California. By age 35 years, the prevalence of moderate/high-intensity smoking in the 1970-1979 birth cohort was 4.6% (95% CI, 3.0%-6.1%) in California and 13.5% (95% CI, 11.8%-15.1%) in the remaining United States.
Conclusions: Between 1965 and 2007, the prevalence of high-intensity smoking decreased greatly in the United States. The greater decline in high-intensity smoking prevalence in California was related to reduced smoking initiation and a probable increase in smoking cessation.