Proximal and distal colorectal cancer resection rates in the United States since widespread screening by colonoscopy

Abstract

Background & aims: Screening decreases colorectal cancer (CRC) incidence and mortality. Colonoscopy has become the most common CRC screening test in the United States, but the degree to which it protects against CRC of the proximal colon is unclear. We examined US trends in rates of resection for proximal vs distal CRC, which reflect CRC incidence, in the context of national CRC screening data, before and since Medicare's 2001 decision to pay for screening colonoscopy.

Methods: We used the Nationwide Inpatient Sample, the largest US all-payer inpatient database, to estimate age-adjusted rates of resection for distal and proximal CRC, from 1993 to 2009, in adults. Temporal trends were analyzed using Joinpoint regression analysis.

Results: The rate of resection for distal CRC decreased from 38.7 per 100,000 persons (95% confidence interval [CI], 35.4-42.0) to 23.2 per 100,000 persons (95% CI, 20.9-25.5) from 1993 to 2009, with annual decreases of 1.2% (95% CI, 0.1%-2.3%) from 1993 to 1999, followed by larger annual decreases of 3.8% (95% CI, 3.3%-4.3%) from 1999 to 2009 (P < .001). In contrast, the rate of resection for proximal CRC decreased from 30.0 per 100,000 persons (95% CI, 27.4-32.5) to 22.7 per 100,000 persons (95% CI, 20.6-24.7) from 1993 to 2009, but significant annual decreases of 3.1% (95% CI, 2.3%-4.0%) occurred only after 2002 (P < .001). Rates of resection for CRC decreased for adults ages 50 years and older, but increased for younger adults.

Conclusions: These findings support the hypothesis that population-level decreases in rates of resection for distal CRC are associated with screening, in general, and that implementation of screening colonoscopy, specifically, might be an important factor that contributes to population-level decreases in rates of resection for proximal CRC.

Figure 1

US colorectal cancer resection rates for all adults. In contrast with distal colorectal cancer resection rates, proximal colon cancer resection rates decreased significantly only after 2002. Curves show the results of Joinpoint analyses. (A) Colorectal cancer overall, (B) distal colorectal cancer, and (C) proximal colon cancer.

Figure 2

US overall colorectal cancer resection rates by age group. In contrast with decreasing rates in adults eligible for screening, increasing rates were observed in younger adults. Curves show the results of Joinpoint analyses. (A) Persons aged ≥76 years, (B) persons aged 50–75 years, (C) persons aged 40–49 years, and (D) persons aged 18–39 years.

Figure 3

US distal and proximal colorectal cancer resection rates by age group. Curves show the results of Joinpoint analyses. (A) Distal colorectal cancer in persons age ≥76 years. (B) Proximal colon cancer in persons age ≥76 years. (C) Distal colorectal cancer in persons age 50–75 years. (D) Proximal colon cancer in persons age 50–75 years. (E) Distal colorectal cancer in persons age 18–39 and 40–49 years. (F) Proximal colon cancer in persons age 18–39 and 40–49 years.

Figure 4

US colorectal cancer resection rates by sex. Curves show the results of Joinpoint analyses. (A) Colorectal cancer overall in women, (B) colorectal cancer overall in men, (C) distal colorectal cancer in women, (D) distal colorectal cancer in men, (E) proximal colon cancer in women, and (F) proximal colon cancer in men.

Figure 5

Trends in colorectal cancer resection rates and use of screening tests. (A) US colorectal cancer resection rates for all adults. Curves show the results of Joinpoint analyses. (B) Data on the prevalence of colorectal cancer screening in adults age 50 years and older from the Behavior Risk Factor Surveillance System. Colonoscopy has been the most common screening test since Medicare’s decision to cover screening colonoscopy in average-risk persons in 2001, and similar decisions by commercial insurers. The American Cancer Society and subspecialty society guidelines endorsed screening colonoscopy in 1997, and the US Preventive Services Task Force endorsed it in 2002. Colonoscopy accounted for most lower endoscopies in the past decade, and sigmoidoscopy accounted for only a small fraction. Proximal colon cancer resection rates began decreasing only in the era of screening colonoscopy.