We read with interest the study by Mohamed Hoesein et al1 (see page 1530), in which they nicely proved that rate of lung function decline is maximum in male heavy smokers with mild or no airflow limitation as documented on spirometry. However, the criteria for categorizing all enrolled patients into COPD have not been specified. In view of the lack of spirometric evidence of airway obstruction and reversibility, it is plausible that some of the subjects may actually be suffering from other airway diseases, notably asthma, which usually do not result in progressive lung function decline. Segregation of such patients by serially testing with spirometry any time during the course of 3 years could have increased the specificity of the study. Moreover, COPD is also known to occur in nonsmokers, predisposed by other risk factors like air pollution,2 genetic predisposition,3 and so forth. Inclusion of nonsmokers might have given a better picture of the association between rate of decline in lung function and stage of COPD.