Abstract: . . .  American Cancer Society, Inc. Page 20 New Cancer Cases: The estimated numbers of new US cancer cases among African Americans in 2003 were calculated by fitting the estimated numbers of cancer cases that occurred each year in the US from 1979 through 1999 to a statistical forecasting model. The esti- mated numbers of US cases from 1979 through 1999 were calculated using cancer incidence from the regions of the US included in the Surveillance, Epidemiology, and End Results (SEER) program of the National Cancer Inst . . .
. . .  adjusting for age, it would be inaccurate to compare the cancer rates of the state of Florida, which has a large elderly population, to that of Alaska, which has a younger population. This is especially true when examining cancer rates, since cancer is generally a disease of older people. Without adjusting for age, it would appear that the cancer rates for Florida are much higher than Alaska. However, once the ages are adjusted, it appears their rates are similar. Starting with this issue of Cancer Facts & Figures for African Americans 2003-2004 , we used the most recent US census (2000) as the baseline for our age-adjustment. This is a change from previous issues and other statistics we have published. Prior to this, our statistics were based on the 1970 census. This change follows federal agencies that publish statistics. The change will also require a recalculation of age-adjusted rates for previous years to allow valid comparison between current and past years. The impact of this change . . .
. . .  process also employs a system for rating strength and consistency of evidence that is similar to that employed by the Agency for Health Care Research and Quality (AHCRQ) and the US Preventive Services Task Force (USPSTF). ©2003, American Cancer Society, Inc. Page 20 New Cancer Cases: The estimated numbers of new US cancer cases among African Americans in 2003 were calculated by fitting the estimated numbers of cancer cases that occurred each year in the US from 1979 through 1999 to a statistical forecasting model. The esti- mated numbers of US cases from 1979 through 1999 were calculated using cancer incidence from the regions of the US included in the Surveillance, Epidemiology, and End Results (SEER) program of the National Cancer Inst . . .
--3000,3,500,2486,51299