We analyzed Lung Health Study (LHS) data to assess the effect of self-reported lower respiratory illnesses resulting in physician visits (LRI) on lung function. Participants were 5,887 smokers aged 35-60 yr, FEV(1)/FVC < 0.70 and FEV(1) of 55-90% predicted. Two-thirds were randomized into an intensive smoking cessation program (SI); one-third were advised only to stop smoking (UC). For 5 yr participants had annual spirometry and questioning regarding LRI. SI had greater rates of smoking cessation than usual care (UC) with fewer LRI (p = 0.0008). Sustained quitters had fewer LRI than continuing smokers (p = 0.0003). In the year LRI occurred, FEV(1) did not change in sustained quitters, but decreased significantly in smokers (p = 0.0001) with some recovery the following year if no LRI occurred. Over 5 yr, LRI had a significant effect on rate of decline of FEV(1) only in smokers. In smokers averaging one LRI/yr over 5 yr there were additional declines in FEV(1) of 7 ml /yr (p = 0.001). Smokers with more than one LRI/yr had greater declines. Chronic bronchitis was associated with increased frequencies of LRI, but did not affect their influence on lung function. Smoking and LRI had an interactive effect on FEV(1) in people with mild COPD, and in smokers frequent LRI may influence the long-term course of the disease.