There has been controversy in the last decade over whether the reported increase in brain tumors reflects a real increase in incidence rates. Incidence data on the full spectrum of brain tumors is lacking in the discussion since current cancer reports in the United States are restricted to malignant tumors. Data on tumors from four population-based cancer registries in the United States were compiled to provide incidence rates of benign and malignant brain tumors and to assess the feasibility of providing these data on a larger scale. A total of 8,070 primary tumors diagnosed from 1985 to 1989 in Connecticut, Massachusetts, Missouri and Utah were obtained. Brain tumors were defined using the International Classification of Diseases for Oncology codes 191.0-191.9, 192.0-192.3, 192.8-192.9 and 194.3-194.4. Stratum-specific incidence rates by location and histology were estimated by sex, age and region. Age-adjusted rates were standardized to the 1970 United States population. An age-adjusted incidence rate of 9.4/10(5) was observed, which reflects a 36% increase in males and a 68% increase in females over the rate based on malignant tumors alone from the Surveillance, Epidemiology and End Results cancer reporting system. Incorporating benign tumors into cancer registry data would increase the reported incidence rates primarily in females and for meningiomas and nerve sheath tumors. This expanded incidence rate represents a substantial improvement in the ability to describe the occurrence of these complex tumors by subtype with a modest increase in overall case registrations for cancer registries. Centralization of data on all brain tumors appears feasible. Variations in histology-specific rates across regions raises questions that need to be addressed about the ascertainment and accuracy of tumor classification. Use of the cancer registration system to improve the reporting of brain tumors in the United States is important to our understanding of the occurrence of these complex tumors and to our ability to conduct large-scale epidemiologic investigations.