Although the information that has been reviewed leaves many questions unanswered, some conclusions can be drawn from available data. (1) Smoking appears to increase the risk of sensitization to certain inhaled antigens encountered in the workplace; however, there is no definite evidence that smoking increases the frequency or intensity of allergy to common aeroallergens in the general population. On average, smokers have higher serum total IgE concentrations and blood eosinophil counts than do nonsmokers, but the mechanisms underlying these alterations are not clear. Analysis of these relationships is complicated by observations suggesting that atopic persons are less likely to become and to remain regular cigarette smokers. (2) Long-term cigarette smoking may be associated with increased nonspecific airway responsiveness, although the magnitude of this effect is relatively small when one adjusts for prechallenge level of pulmonary function. This effect of smoking may be more pronounced in atopic persons. (3) Atopy, as assessed by skin testing and serum IgE concentrations, is associated with asthma, nonspecific airway hyperresponsiveness, and reduced pulmonary function level in population data. However, there is no clear evidence that atopy is a risk factor for irreversible airflow obstruction in persons without asthma. Population data do not indicate how much of the reduction in pulmonary function associated with atopy and asthma is potentially reversible. (4) Blood eosinophil count appears inversely related to the level of pulmonary function and directly related to the rate of decline of pulmonary function among nonsmokers. Reports vary concerning whether the relationship of eosinophil count to level of pulmonary function remains after excluding subjects with diagnosed asthma. This relationship may be determined largely by a clinically distinguishable subset of nonsmokers with "asthmatic bronchitis." Presumably, these observations reflect an adverse impact of eosinophilic inflammation in the airways or lung parenchyma. It is not clear whether this represents an allergic response because skin-test reactivity to common aeroallergens and serum total IgE concentration do not show similar relationships to reduced level and rapid decline of pulmonary function. (5) Among smokers, nonspecific airway hyperresponsiveness appears to be associated with an accelerated longitudinal decline of pulmonary function, although most studies indicating this association are limited by either a retrospective design or lack of adjustment for prechallenge level of pulmonary function.(ABSTRACT TRUNCATED AT 400 WORDS)