Objective: To assess the hypothesis that higher incidence, not lower severity threshold for admission, explains differences in asthma hospitalization rates (AHRs) between inner city (IC) and suburban (SUB) children. Design/Methods: Zip codes defined residence as IC, other urban (OU), or SUB for the 198,593 children (<19 yr) dwelling in the Rochester, NY area. All 2,028 asthma hospitalizations between 1991 and 1995 for these children were analyzed. Severity, defined as mild (≥90) or severe(<90) by the worst SaO2 on the day of admission, was examined by hospital record review for random samples of 165 SUB and 169 IC asthma admits. Results: AHR is expressed as admits/1000 child-years. Differences are expressed as odds ratios (OR) with 95% CI. Overall AHR was 2.04. AHRs within the community were ordered: SUB (1.05) < OU (2.99) < IC (5.21). The proportion of admits with a low severity threshold for admission (mild episodes) was slightly greater for IC children (55.2% vs 45.6%, P=.08). For both severe and mild episodes, however, AHRs were much greater for IC children(mild: OR=4.1 [3.6-4.7]; severe: OR=6.0 [5.2-6.9]). Conclusion: The marked socioeconomic disparity in Rochester area AHRs is largely attributable to higher incidence of acute asthma attacks among IC children.