Background. Studies indicate that the incidence of primary malignant brain tumors in the elderly is increasing, but this may reflect increased case ascertainment due to the introduction of computed tomography (CT) scanning. In addition, tumor histology was not available in these studies. This study was conducted to determine whether the incidence of primary malignant brain tumors in the elderly in Florida is increasing and to verify the histology of these increases.
Methods. Using the number of primary malignant brain tumors reported to the Florida Cancer Data System (FCDS) when CT scanning was available, incidence rates per 100,000 people were calculated. Incidence density ratios (IDRs) were calculated for 1986–1989 relative to 1981‐1984.
Results. Tumor incidence at ages 20–64 years increased from 5.7 in 1981–1984 to 5.9 in 1986–1989, with an IDR of 1.05 (not significant). The incidence in those aged 65 years or older rose from 14.8 to 18.3, with an IDR of 1.23 (P < 0.001). The increase was 15% in those aged 65–69, 16% in those 70–74 years, 30% in those 75–79 years, 36% in those 80–84 years, and 254% in those 85 years or older. Indicence density ratios in those aged 65 years or older were 0.92 (not significant) for astrocytoma, 2.7 (p < 0.001) for anaplastic astrocytoma, 1.32 (p < 0.001) for glioblastoma and 3.56 (p < 0.001) for lymphoma. In those aged 65+ the incidence of all cancers rose 7.6% (not significant), and the incidence of pancreatic cancer (another neoplasm that requires CT scanning for diagnosis) rose 0.34% (not significant).
Conclusions. The incidence of primary brain tumors in elderly Floridians has increased. This increase is independent of increased case ascertainment associated with the introduction of CT scanning and independent of a general increase of all cancers. The rise in brain tumor incidence is observed in anaplastic astrocytoma, glioblastoma, and lymphoma but not astrocytoma. This study confirms the increase is histologically specific and not due to increased case ascertainment. Further investigation into the etiology of this increase is warranted. Cancer 1995; 76:1634–42.