Changing trends in the incidence, stage, survival, and screen-detection of colorectal cancer: a population-based...

Colorectal cancer (CRC) screening has been advocated increasingly during the past 2 decades, but there is little direct evidence that it has affected cancer incidence or presentation at the population level. This study assessed concurrent trends in CRC incidence, presentation, survival, mortality, and polypectomies in Olmsted County, Minnesota.

Longitudinal observational study was conducted of all Olmsted County residents with colorectal adenocarcinoma first diagnosed in 1980-1999.

Altogether, 889 invasive CRCs were diagnosed among Olmsted County residents from 1980 through 1999. Annual age- and sex-adjusted CRC incidence rates decreased significantly during this period (P = .02) with a 23% decline from 60 per 100,000 in 1980-1984 to 46 per 100,000 in 1995-1999. This was primarily accounted for by a 40% reduction in left-sided CRC (P < .001). The incidence of right-sided CRC remained unchanged, but the proportion of right-sided CRC increased from 46% to 58%. Cancer stage shifted downward, and survival improved over time. The overall proportion of screen-detected CRC rose from 8% to 17%. Annual adjusted adenomatous polypectomy rates increased dramatically from 86 to 320 per 100,000 (P < .001).

The incidence of CRC in Olmsted County has declined in recent years. An impact of screening is suggested by the observed upward trend in screen-detected cases, a favorable stage shift, and a concurrent rise in polypectomy rates. However, the incidence of right-sided cancer remains unaltered and might be less affected by historically used screening interventions. Most CRCs still present symptomatically, and more effective population screening is needed.