STUDY OBJECTIVE: To assess the short term relationship between air pollution and the daily number of emergency hospital admissions for respiratory disease. DESIGN: Data were analysed using autoregressive Poisson regression allowing for overdispersion and controlling for possible confounding factors such as seasonal and other chronological variables, meteorological factors, and influenza epidemics. SETTING: The two major cities in The Netherlands-Amsterdam (694,700 inhabitants) and Rotterdam (576,200 inhabitants). PARTICIPANTS AND MEASUREMENTS: Emergency hospital admissions for respiratory diseases, registered on a daily basis by the National Medical Registration, for the period 1977-89 were used. ICD-9 codes included were: respiratory (460-519), chronic obstructive pulmonary disease (490-492, 494, 496), and asthma (493). The mean (range) of the total daily number of admissions for these three classifications were as follows: 6.70 (0-23), 1.74 (0-9) and 1.13 (0-7) respectively in Amsterdam and 4.79 (0-19), 1.57 (0-9), and 0.53 (0-5) in Rotterdam. Air pollution measurements were provided by the National Institute of Public Health and Environmental Protection. In The Netherlands, air pollution is at a low to moderate ("summer type") or a low ("winter type") level. The levels in Amsterdam and Rotterdam did not differ much for the "summer type". For 1977-89 the mean (range) values of ozone (O3), the "summer type" pollutant (O3-8 h), were 86 (0-252) micrograms/m3 in Amsterdam and 82 (0-286) micrograms/m3 in Rotterdam. The mean (range) of the values "winter type", pollutant, sulphur dioxide (SO2-24 h), were 38 (0-381) micrograms/m3 in Amsterdam and 50 (1-379) micrograms/m3 in Rotterdam. For black smoke (BS-24 h), values were 14 (1-84) micrograms/m3 and 28 (1-144) micrograms/m3 respectively (1986-89). MAIN RESULTS: Ozone had a non-significant positive effect on the number of respiratory emergency admissions in summer in people aged > or = 65 years (relative risk for a 100 micrograms/m3 increase in O3-8 h of 1.127 (0.983, 1.292) in Amsterdam and a significant positive effect of 1.344 (1.097, 1.647) in 1977-81 in Rotterdam). Sulphur dioxide did not show any clear effects; in Amsterdam a significant negative effect was even found. The same was true for nitrogen dioxide in Amsterdam; in Rotterdam, however, nitrogen dioxide showed non-significant positive effects (RR 0.965, 1.342). Black smoke did not show any clear effects in Amsterdam; in Rotterdam it was positively but not significantly related to the number of admissions. CONCLUSIONS: The results show that the relation between short term air pollution and emergency hospital admissions is not always consistent at these rather low levels of daily hospital admissions and of air pollution.