The 13th World Conference on Tobacco OR Health

Objective: To assess whether observed lower smoking rates in elderly is explained as cohort effect or differential quitting rates.

Methods: Maryland Cancer Survey 2004, was a population-based, random digit-dial, computerized assisted telephone interview of 5,004 persons 40 years and older. Current smoking rate was compared between those 40-64 years and 65+ years. Logistic regression was used to adjust for possible confounders.

Results: Current smoking rates were 20% in the 40-64 year old cohort versus 8% of those 65+. The differences of the current smoking rates between the two age groups persisted after adjusting for possible confounding variables such as education, income, race, marital status, and sex (OR= 2.8, 95%CI= 2.1,3.9). The rate of “ever smoked” was not different between the two groups (50% vs. 52%). Among “ever smokers” the 40-64 year olds were less likely to have quit smoking (61%) than those 65+ (84.2%) (OR= 0.3, 95%CI= 0.2,0.4). We found an interaction between gender and age. Men 40-64 years were significantly less likely than older men to have “ever smoked” (OR= 0.6, 95%CI= 0.5,0.8) but also less likely to have quit; women aged 40-65 were slightly more likely to have “ever smoked” than women 65+ (OR= 1.1, 95%CI= 0.9,1.3).

Conclusion: The lower current smoking rates observed in elderly men seem to be primarily due to increased quitting in older age groups and not due to a cohort effect. However, in women, the lower rate in those 65+ could be explained by both increased quitting rate and a possible cohort effect.