In a representative sample of 72 sentinel practices in Lower Saxony and 61 sentinel practices in Saxony-Anhalt, physicians were requested to record all consultations of asthmatic children (aged up to 8 years) over a period of 6 and 3 months, respectively. Children were classified as asymptomatic or newly symptomatic at each separate contact. Asymptomatic contacts with respiratory tract infections were excluded. From measurements of fixed air monitoring stations, the exposure to air pollutants (NO 2, NO, SO 2, ozone, particulates) at the place of the physician's practice (Lower Saxony) or at the child's residence (Saxony-Anhalt) was estimated, using a specific interpolation method (inverse distance weighting). Associations between exposure to particular air pollutants and the presence of asthmatic symptoms at the time of the consultation were determined using logistic regression. Temperature, month of the consultation, region, day of the week, first or repeated contact in the recording period were included as covariates in the multivariate analyses to adjust for potential confounding. NO 2 exposure was significantly and positively associated with the occurrence of asthmatic symptoms (p = 0.007). Exposure to other air pollutants did not significantly differ between asymptomatic and symptomatic contacts. A possible causal relationship between NO 2 exposure and the occurrence of asthmatic symptoms in children suffering from asthma is in accordance with other studies. Our study adds data from routine office based outpatient care to independent evidence on the association between NO 2 exposure and asthma.