November 15, 2007 / Volume 110 / Number 10
Prevalence estimates for screening and risk fac- tors were calculated using data from the BRFSS aggregated over the 6-year period 1999 through 2005. Not all years were available for each BRFSS-derived risk factor, because some questions were not asked every year. Prevalence estimates from BRFSS were age-adjusted to the 2000 U.S. standard population and sample-weighted using SUDAAN software to account for the complex BRFSS sampling design.43 Prevalence estimates for cancer risk factors and screening were suppressed if they were based on < 50 observations. Prevalence estimates of socio- economic characteristics from U.S. Census were not age-adjusted. Confidence intervals are provided for all prevalence estimates; however, they do not pro- vide a basis for significance testing of regional differ- ences.44
Regional, site-specific comparisons of incidence rates between AI/AN and NHW populations are shown by rate ratios (RR) with 95% CIs. In describing rate comparisons for AI/AN and NHW populations, the terms ‘‘higher’’ or ‘‘lower’’ were used when the RRs differed significantly from 1.0 (P < .05). Thus, when the RR was < 1.0, the rate among AI/AN was
was > 1.0, the rate among AI/AN was higher than that among NHW. Otherwise, the RRs were described as comparable. Relative percents (R%) were cal- culated to facilitate comparisons of distributions of age-adjusted, stage-specific incidence rates between NHW and AI/AN populations across IHS regions.
RESULTS Update on Long-term Incidence Trends, 19752004 For all populations, both sexes, and all cancer sites combined, incidence rates increased from 1975 through 1992 and then decreased (0.3% per year) from 1992 through 2004 (Table 2). For men, the rates increased from 1975 through 1992, decreased from 1992 through 1995, and subsequently stabilized from 1995 through 2004. For women, rates increased from 1979 through 1986, increased more slowly from 1986 through 1999, and stabilized from 1999 through 2004.
Among men, prostate cancer incidence rates sta- bilized from 1995 through 2004 after the steep increase reported from 1988 through 1992 and the subsequent decline from 1992 through 1995. Increases in cancer incidence were noted from 1975 through 2004 for myeloma and for cancers of the kidney and renal pelvis (kidney), liver and intrahe- patic bile ducts (liver), and esophagus. Declining trends in incidence were noted for the most recent
joinpoint segments for cancers of the lung and bronchus (lung), colon and rectum, oral cavity and pharynx (oral cavity), and stomach. Rates were stable for the most recent joinpoint segments for the remaining top 15 cancers in men.
Among women, breast cancer incidence rates decreased for the period 2001 through 2004 by 3.5% per year, reversing a long-term increase that began in the early 1980s. Lung cancer rates increased from 1975 through 1998 and stabilized from 1998 through 2004. Colorectal cancer rates decreased from 1998 through 2004. Rates also declined for cancers of the cervix and stomach from 1975 through 2004 and for cancers of the corpus uteri and uterus not otherwise specified (uterus), ovary, and oral cavity in the most recent joinpoint segments. Increasing trends were noted from 1975 through 2004 for melanoma, non- Hodgkin lymphoma (NHL), leukemia, and cancers of the bladder and kidney. Increasing rates of thyroid cancer were noted from 1980 through 1995, with the rate of increase nearly tripling from 1995 through 2004. Pancreatic cancer incidence rates stabilized for the most recent joinpoint segment.
Update on Long-term Mortality Trends, 19752004 Cancer death rates for all races/ethnicities and both sexes combined decreased by 2.1% per year from 2002 to 2004 compared with 1.1% per year from 1993 to 2002 (Table 3). The decreases involved both men and women, but they were greater in men than in women from 2002 through 2004 (2.6% per year in men and 1.8% per year in women). Accelerated declines in cancer mortality rates largely reflect more rapid decreases in the death rates from leading causes of cancer mortality: lung and prostate cancer in men, breast cancer in women, and colon and rec- tum cancer in both men and women.
During the last joinpoint segment, rates were declining for 12 of the 15 most common causes of cancer death in men (lung, prostate, colon and rec- tum, pancreas, leukemia, NHL, bladder, kidney, stomach, brain, myeloma, and oral cavity) (Table 4). Melanoma rates remained stable from 1990 through 2004. The rates for only 2 sites—cancer of the liver and esophagus—increased (1.8% and 0.4% per year, respectively).
Similarly, rates declined for 10 of the 15 most common causes of cancer death in women (breast, colon and rectum, NHL, leukemia, brain, myeloma, stomach, kidney, cervix, and bladder). Death rates for the most recent joinpoint segments were stable in women for cancers of the pancreas, ovary, and