Longitudinal analysis of the effects of smoking onset and cessation on pulmonary function

We examined the effects of smoking onset and cessation on FEV1 in participants in the Tucson Epidemiological Study of Airways Obstructive Disease 18 yr of age or older who reported a change in smoking habits during 17 yr of follow-up. Subjects were classified as "new quitters" (n = 288) and "new starters" (n = 45) at each survey on the basis of questionnaire responses concerning current smoking habits. The pulmonary function data were analyzed using a mixed longitudinal or random effects model (REM). We compared FEV1 before and after changing smoking habits in the same subjects while adjusting for important covariables such as height, pack-years, and respiratory symptoms and diseases. The results for smoking cessation showed a beneficial effect related to quitting that was largest for younger subjects and decreased linearly with age. For women, quitting resulted in an improvement in FEV1 of 4.3% at 20 yr of age which decreased to 2.5% by 80 yr of age. Men had an improvement of only 1.2% at 20 yr of age and no improvement at 80 yr of age. After excluding subjects with low initial function (FEV1/FVC in the lowest quartile) and those with only a single observation after quitting, men (n = 70) showed a significantly higher FEV1 improvement of 4% at 20 yr of age that decreased linearly to zero by age 80 yr. In contrast, women in this subgroup (n = 89) had nearly the same degree of improvement at all ages. The onset results showed mean FEV1 values after starting smoking that were higher at the younger ages (< 23 yr) and that decreased more rapidly with age, compared with nonsmoking mean estimates, in both sexes.