LAUSR

810 Cancer March 1, 2021 individually. However, because the test did not reject this theory for Black women (P = .10), age groups were analyzed together. When the P value was <.05, the results were deemed statistically significant; otherwise, they were considered nonsignificant. Rate changes with absolute APC values of 0.0 to 0.9 were deemed slow, those with absolute APC values of 1.0 to 1.9 were deemed moderate, those with APC values of 2.0 to 3.0 were deemed rapid, and those with APC values >7.0 were deemed extremely rapid. Figure 1 displays results of Joinpoint regression analyses. For White women aged 50 to 64 years, a significant, rapid decline in the rate for 2000 through 2004 of −2.9 APC was preceded by a nonsignificant, slow rate decline of −0.2 APC, followed by a period (2004-2017) without a rate change (0.0 APC). For White women aged 65 to 79 years, the year 1992 had the highest recorded rate, at 824.5 diagnoses per 100,000 women-years. A nonsignificant, rapid decline in the rate for 1992 through 1994 of −2.5 APC was followed by a nonsignificant, moderate rise in the rate for 1994 through 1998 at 1.0 APC. A significant, rapid decline followed in the rate for 1998 through 2004 at 2.3 APC. A nonsignificant, slow rise in the rate for 2004 through 2013 of 0.4 APC was followed by a nonsignificant, slow rate decline of −0.9 APC. The findings suggest that, for White women aged <80 years who are insured by Medicare, rates began declining before 2000, and possibly as early as 1992. The Women’s Health Initiative; Hormone Replacement Therapy; and Surveillance, Epidemiology, and End Results Data