SEARCH

SEARCH BY CITATION

Keywords:

  • colon carcinoma;
  • rectal carcinoma;
  • incidence;
  • race;
  • subsite;
  • disease stage

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

BACKGROUND.

Cancers of the colon and rectum are the third most common malignancy among males and females in the United States, although incidence and mortality have declined in recent years. We evaluated recent trends in colorectal cancer incidence in the United States by subsite and stage at diagnosis.

METHODS.

Data for this analysis included all cases of colorectal cancer diagnosed between 1992 and 2001 and reported to the National Cancer Institute's Surveillance, Epidemiology, and End Results (SEER) Program. Incidence rates were stratified by gender, race/ethnicity, anatomic subsite, stage at diagnosis, and SEER registry. Trends in incidence over time were measured using the estimated annual percentage change.

RESULTS.

The study population included 95,539 males and 93,329 females with colorectal cancer. For all 12 SEER registries combined, incidence declined between 1992 and 2001 by 1.2% per year among males and 0.7% per year (not statistically significant) among females. Rates for non-Hispanic whites declined by an average of 1.3% per year for males and 0.6% per year for females. Overall rates for black, Asian/Pacific Islander, and Hispanic males and females did not change significantly except for a 0.8% decline among Asian/Pacific Islander males. Declines in rates among males and females were most pronounced for tumors of the sigmoid colon.

CONCLUSIONS.

Colorectal cancer rates decreased in the United States during the 1990s. Decreases were most pronounced among males, among non-Hispanic whites, and for tumors of the sigmoid colon. These reductions are probably dueto the increased use of screening. Cancer 2006. © 2006 American Cancer Society.

Cancers of the colon and rectum are the third most common malignancy and the third most frequent cause of cancer death among males and females in the United States, although both incidence and mortality have declined in recent years.1 Researchers debate whether the changes in colorectal cancer rates can be attributed to improved screening or a change in risk factors.2 Utilization of screening has increased during the past 2 decades,3 but some evidence also indicates improvements in diet among Americans.4 Changes in screening utilization and diet are likely to vary by race/ethnicity and geographic residence.

The annual age-adjusted incidence of colorectal cancer in the United States rose after World War II, peaked in 1985, and declined in incidence thereafter.2 In this article, we evaluate trends in colorectal cancer incidence between 1992 and 2001 in the United States by subsite and stage at diagnosis with the use of data from the National Cancer Institute's Surveillance, Epidemiology, and End Results (SEER) Program. The purpose of this analysis was to examine trends in recent SEER data, with a focus on patterns of colorectal cancer incidence by gender, race/ethnicity, and geographic area of residence.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Study Population and Sources of Data

Data for this analysis were obtained from the SEER Program, a population-based cancer registration program that began collecting data on cases in defined geographic areas in 1973.5 The 12 SEER registries in this study currently collect and publish cancer incidence and survival data covering ˜14% of the U.S. population. SEER data include patients' demographic characteristics, tumor characteristics, and initial therapy. The primary sources of data are hospital medical records, pathology and radiotherapy reports, outpatient surgical center records, and death certificates.

In this analysis, we included all cases of colorectal cancer diagnosed between 1992 and 2001 and reported by SEER registries in 5 states (Connecticut, Iowa, Hawaii, New Mexico, and Utah), 6 metropolitan areas (Atlanta, Detroit, San Francisco-Oakland, Seattle-Puget Sound, Los Angeles, and San Jose-Monterey), and the Alaska Native population. Data from the contiguous San Francisco-Oakland and San Jose-Monterey registries were subsequently combined and presented as the San Francisco Bay Area. Estimates in this study were based on 2 SEER public-use files, both of which contain incidence data for 1992–2001 from the November 2003 submission: the Expanded Races file6 and the Hispanics file.7

Definition of Variables

Incidence rates were stratified by gender, race/ethnicity, anatomic subsite, stage at diagnosis, and SEER registry. Age-adjusted incidence rates and trends for colorectal cancer were calculated for 4 racial/ethnic groups: non-Hispanic (NH) white, black, Asian/Pacific Islander (PI), and Hispanic. Rates and trends for black and Asian/PI patients were based on the Expanded Races file, whereas trends for NH whites and Hispanics were based on the Hispanics file. Among persons of Hispanic ethnicity, 98.1% were white, 1.7% were nonwhite, and 0.2% were of unknown race. Although information on Hispanic ethnicity was not available for either black or Asian/PI persons, these 2 groups represent a large portion of the 1.7% nonwhite Hispanic population. The “all races” category included persons of other or unknown race/ethnicity.

Stage at diagnosis for all cases was coded according to SEER Summary Stage guidelines.8 In situ tumors were defined as noninvasive tumors that had not penetrated the basement membrane; localized tumors were those confined entirely to the organ of origin; regional tumors were those that extended into surrounding organs and tissues (or regional lymph nodes); and distant tumors were those that had spread to remote organs or lymph nodes. Regional and distant stages were subsequently combined into a single group to represent cases with a “late-stage” diagnosis. Tumors of unknown stage were included in the “all stages” category only.

Anatomic site for each tumor diagnosed after 2000 was coded according to the International Classification of Diseases for Oncology, 3rd edn. (ICD-O3).9 Cases diagnosed before 2001, which had been coded by ICD-O2,10 were converted to ICD-O3 codes. Cases with an anatomic site coded C18.0 through C20.9 and C26.0 were included in this analysis; cases with a nonspecific site (C18.8, C18.9, C26.0) or site in the appendix (C18.1) were included in the “all sites” category only. Anatomic sites were grouped into 3 subsites: proximal colon (C18.0-C18.6, corresponding to cecum, ascending colon, hepatic flexure, transverse colon, splenic flexure, and descending colon), sigmoid colon (C18.7), and rectum (C19.9-C20.9, or rectum and rectosigmoid junction). The flexible sigmoidoscope, used most often for screening of the colon, rarely reaches above the sigmoid colon11; thus, all sites above the sigmoid colon were included with the proximal colon to evaluate the effect of screening on incidence.

Statistical Analysis

Incidence was calculated only for groups with at least 16 cases to provide stable rates and to protect confidentiality, and rates were age-adjusted to the 2000 U.S. standard population. Trends in incidence over time were measured as the estimated annual percentage change (EAPC), calculated by fitting a linear regression to the natural logarithm of the age-adjusted annual rates (r), with year of diagnosis as the regressor variable: ln r = m(year) + b. From m, the slope of the regression line, the EAPC was calculated as follows: EAPC = 100 × (em − 1).12 A positive or negative EAPC corresponds to an increasing or decreasing trend, respectively. We tested the hypothesis that the EAPC is equal to zero under theassumption that α = 0.05 and assuming that rates changed at a constant rate during the 10-year period of this study. All data analyses in this study were performed with SEER*Stat version 6.1.4 software (National Cancer Institute Surveillance Research Program; www.seer.cancer.gov/seerstat).

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

The study population of 188,868 patients with colorectal cancer included 95,539 males and 93,329 females. For all SEER registries combined, colorectal cancer rates declined between 1992 and 2001 by 1.2% per year among males and by a nonsignificant 0.7% per year among females. Results by gender, age, race/ethnicity, SEER registry, stage at diagnosis, and tumor subsite are shown in Table 1. The decline was most pronounced for males and females aged 65years and older. Colorectal cancer rates increased for males and females younger than 50 years of age, although the results were not statistically significant. Rates for NH whites declined by an average of 1.3% per year for males and 0.6% per year for females. Rates for black males and females, Asian/PI males and females, and Hispanic females declined by less than 1% per year, and these results were not statistically significant except among Asian/PI males. A small increase was observed among Hispanic males, but these results were nonsignificant. Decreased rates were observed in geographic regions covered by most SEER registries, with the exception of an increase among Alaska natives, but the magnitude of the decline was greatest in the San Francisco Bay Area. The decreases were largest for in situ tumors and for those located in the sigmoid colon.

Table 1. Number of Cases and Estimated Annual Percent Change (EAPC) in the Age-Adjusted Incidence Rates of Cancer of the Colon and Rectum, by Sex, Age at Diagnosis, Race/Ethnicity, Registry, Stage at Diagnosis, and Tumor Subsite, SEER* Program, 1992–2001
CharacteristicMalesFemales
No. of casesEAPCNo. of casesEAPC
  • *

    SEER registries cover ˜14% of the U.S. population.

  • Non-Hispanic whites and Hispanic categories are mutually exclusive. Black and Asian/Pacific Islander persons may be of Hispanic ethnicity.

  • “All race/ethnicities” category includes persons of unknown and other race/ethnicities.

  • §

    “All stages” category includes unstaged cases.

  • “All sites” category includes appendix and large intestine, not otherwise specified.

  • The estimated annual percent change is significantly different from zero (P < 0.05).

Age at diagnosis (yr)
 0–4978030.868931.4
 50–6424259−0.518130−0.8
 >6563477−1.668306−0.9
 All ages95539−1.293329−0.7
Race/Ethnicity
 Non-Hispanic white71687−1.370778−0.6
 Black8349−0.59256−0.1
 Asian/Pacific Islander8200−0.86886−0.6
 Hispanic63410.65553−0.1
 All race/ethnicities95539−1.293329−0.7
SEER registry
 Alaska Natives2621.82694.4
 Atlanta4793−1.05178−2.0
 Connecticut11608−0.9117400.0
 Detroit11977−1.211959−0.2
 Hawaii3912−0.52956−1.4
 Iowa10698−0.311228−0.4
 Los Angeles20329−1.219762−0.1
 New Mexico3785−0.73298−0.7
 San Francisco Bay Area15353−2.214866−1.9
 Seattle/Puget Sound9479−1.39002−0.9
 Utah3343−0.830710.1
 All registries95539−1.293329−0.7
Stage at diagnosis
 In situ5948−2.84907−3.4
 Invasive84444−0.882580−0.3
 Localized35614−0.3335170.4
 Regional/Distant48830−1.249063−0.7
 All stages§95539−1.293329−0.7
Tumor subsite
 Rectum29358−0.723161−0.7
 Sigmoid colon22782−2.420156−1.9
 Proximal colon39457−0.945622−0.1
 All sites95539−1.293329−0.7

Trends by Subsite and Stage

Trends by subsite and stage (all race and age groups combined) are shown in Figures 1–3 for males and females, and EAPC values are provided in Table 2. Declines in rates among males and females were most pronounced for tumors of the sigmoid colon, and smaller declines were observed for the proximal colon and rectum. Declines were greatest for in situ tumors at each anatomic subsite; these decreases averaged ˜3% per year among both males and females.

thumbnail image

Figure 1. Proximal colon cancer age-adjusted incidence rates by stage at diagnosis in (A) males and (B) females, SEER Program, 1992–2001. SEER registries cover ˜14% of the U.S. population. Rates in the figure are per 100,000 persons and are age-adjusted to the 2000 U.S. standard population. The “all stages” category includes unstaged cases. Reg/Distant indicates regional/distant disease.

Download figure to PowerPoint

thumbnail image

Figure 2. Sigmoid colon cancer age-adjusted incidence rates by stage at diagnosis in (A) males and (B) females, SEER Program, 1992–2001. SEER registries cover ˜14% of the U.S. population. Rates in the figure are per 100,000 persons and are age-adjusted to the 2000 U.S. standard population. The “all stages” category includes unstaged cases. Reg/Distant indicates regional/distant disease.

Download figure to PowerPoint

thumbnail image

Figure 3. Rectal cancer age-adjustied incidence rates by stage at diagnosis in (A) males and (B) females, SEER Program, 1992–2001. SEER registries cover ˜14% of the U.S. population. Rates in the figure are per 100,000 persons and are age-adjusted to the 2000 U.S. standard population. The “all stages” category includes unstaged cases. Reg/Distant indicates regional/distant disease.

Download figure to PowerPoint

Table 2. Estimated Annual Percent Change in the Age-Adjusted Incidence Rates of Cancer of the Colon and Rectum, by Subsite, Stage at Diagnosis, Age at Diagnosis, and Sex, SEER* Program, 1992–2001
SubsiteStage at diagnosisMalesFemales
0–4950–64>65All Ages0–4950–64>65All ages
  • *

    SEER registries cover ˜14% of the U.S. population.

  • “All stages” category includes unstaged cases.

  • “All sites” category includes appendix and large intestine, not otherwise specified.

  • §

    The estimated annual percent change is significantly different from zero (P < 0.05).

RectumIn situ−2.8−0.5−2.4−2.0−0.2−3.6−4.8§*−4.1§
Localized4.5§2.2§−0.40.65.5§1.9−1.40.2
Regional/Distant2.8§−0.2−2.0§−1.1§2.9§−1.0−1.0§−0.5
All stages2.5§0.5−1.6§−0.73.6§−0.1−1.7§−0.7
Sigmoid colonIn situ−0.8−3.5−4.0§−3.8§−2.8−3.9−2.5−3.1
Localized1.6−1.7§−2.7§−2.3§3.9§−1.5−1.6§−1.2§
Regional/Distant0.3−1.6§−2.4§−2.0§0.2−1.3−2.9§−2.2§
All stages0.3−1.9§−2.7§−2.4§1.0−1.6§−2.3§−1.9§
Proximal colonIn situ−9.2§−0.5−3.5§−3.1§−3.0−3.9−3.0§−3.2§
Localized−0.70.80.10.2§0.60.91.6§1.4§
Regional/Distant−0.2−1.1§−1.1§−1.1§−1.0−1.3§−0.3−0.5
All stages−0.8−0.6−0.9§−0.9§−0.6−0.80.1−0.1
All sitesIn situ−3.9−1.7−3.1§−2.8§−2.2−3.8§−3.3§−3.4§
Localized2.3§0.7−0.8−0.33.4§0.50.10.4
Regional/Distant1.2−0.8§−1.6§−1.2§0.9−1.1§−0.8§−0.7§
All stages0.8−0.5§−1.6§−1.2§1.4−0.8−0.9§−0.7

Trends by Subsite and Age

Trends by subsite, stage at diagnosis, age at diagnosis, and gender for all race groups combined are shown in Table 2. Overall, the incidence of tumors diagnosed at regional or distant stages decreased significantly among persons aged 50 years and older, but not among those younger than 50 years. For both males and females, the incidence of in situ tumors also declined at all subsites and among the 3 age groups examined in this study. In the younger age group (≤49 years old), overall patterns for males and females were similar. In this age group, the incidence of rectal cancer increased significantly during the 10-year period. Except for in situ tumors, this increase was evident at all stages at diagnosis. Whereas proximal colon cancers decreased somewhat during the period (albeit not significantly), the incidence of sigmoid colon cancers increased, particularly for localized tumors.

Among persons diagnosed between 50 and 64 years of age, the direction (if not the magnitude) of estimated changes was also similar for males and females. In this age group, the incidence of localized rectal cancer increased ˜2% per year, although the result was statistically significant only for males. The incidence of sigmoid colon cancer declined significantly (EAPC for all stages combined: −1.9% for males and −1.6% for females). Regardless of statistical significance, the decline in sigmoid colon cancer was detected in tumors at all stages at diagnosis. The incidence of late-stage proximal colon cancer also declined significantly among males and females (EAPC = −1.1% and −1.3%, respectively), although localized tumors at this site increased by less than 1% per year (a nonsignificant change).

Among persons aged 65 years and older, we observed an overall decline in the incidence of colorectal cancer for tumors in all 3 subsites and at all stages at diagnosis. The only significant exception to this finding was for localized tumors of the proximal colon, which increased by 1.6% per year among females and remained unchanged among males.

Trends by Gender and Race/Ethnicity

Males

EAPC values by subsite, stage at diagnosis, race/ethnicity, and gender are provided in Table 3. For NH white males, the incidence of colorectal cancer declined for almost all subsites and stages at diagnosis, although not all results were statistically significant. The decreases were most pronounced for tumors of the sigmoid colon, and rates of both localized and more advanced tumors of the sigmoid declined nearly 3% per year during this period. Declines of 3%–4% for in situ tumors were observed at every subsite.

Table 3. Estimated Annual Percent Change in the Age-Adjusted Incidence Rates of Cancer of the Colon and Rectum, by Sex, Subsite, Stage at Diagnosis, and Race/Ethnicity, SEER* Program, 1992–2001
SubsiteStage at DiagnosisMalesFemales
Non-Hispanic WhiteBlackAsian/Pacific IslanderHispanicNon-Hispanic WhiteBlackAsian/Pacific IslanderHispanic
  • *

    SEER registries cover ˜14% of the U.S. population.

  • Non-Hispanic whites and Hispanic categories are mutually exclusive. Black and Asian/Pacific Islander persons may be of Hispanic ethnicity.

  • “All stages” category includes unstaged cases.

  • §

    “All sites” category includes appendix and large intestine, not otherwise specified.

  • The estimated annual percent change is significantly different from zero (P < 0.05).

RectumIn situ−3.13.2−2.25.3−4.1−1.0−6.4−1.8
Localized0.71.10.8−1.20.40.60.4−0.6
Regional/Distant−1.2−1.2−0.70.1−0.81.6−0.40.2
All stages−0.8−0.6−0.5−0.2−0.70.4−1.0−0.3
Sigmoid colonIn situ−4.2−0.3−4.1−0.8−3.1−5.1−1.6−6.5
Localized−2.5−1.3−2.81.3−0.9−1.9−1.61.0
Regional/Distant−2.5−0.8−0.80.2−2.4−1.1−1.0−3.7
All stages−2.7−0.8−2.00.7−1.8−1.8−1.6−2.2
Proximal colonIn situ−4.00.1−1.8−0.2−3.1−0.3−6.6−3.4
Localized−0.4−0.1−0.62.91.71.31.12.8
Regional/Distant−1.20.10.80.1−0.6−0.11.80.1
All stages−0.9−0.40.21.00.00.20.80.6
All sites§In situ−3.71.3−2.92.2−3.4−1.3−4.3−4.2
Localized−0.20.0−0.80.90.70.4−0.11.2
Regional/Distant−1.4−0.4−0.10.4−0.90.30.4−0.4
All stages−1.3−0.5−0.80.6−0.6−0.1−0.6−0.1

Colorectal malignancies also declined among Asian/PI males, although to a lesser extent than among NH white males. As in the NH whites, declines in Asian/PI males were greatest for tumors of the sigmoid colon and for in situ tumors at every subsite. The only exceptions were nonsignificant increases of 0.8% per year for late-stage tumors of the proximal colon and for localized tumors of the rectum.

Unlike NH whites and Asian/PIs, black males experienced no significant decreases in colorectal cancer incidence for any subsite or stage group. A slight decline in tumors of the sigmoid colon and late-stage rectal tumors was suggested, but these results were not statistically significant.

Trends among Hispanic males were substantially different from those in the other racial/ethnic groups. The incidence of localized tumors of the sigmoid and proximal colon increased (by 1.3% and 2.9% per year, respectively), as did in situ tumors of the rectum. For this ethnic group, the incidence of localized rectal tumors and in situ tumors of the sigmoid colon decreased, but the incidence of late-stage tumors at all subsites remained unchanged. None of these results were statistically significant.

Females

For NH white females, declines were observed for tumors of the rectum and sigmoid colon, but the incidence of localized tumors of the proximal colon increased by 1.7% per year (Table 3). NH white females experienced small, nonsignificant decreases in late-stage tumors of the proximal colon and rectum (−0.6% and −0.8% per year, respectively), and the incidence of late-stage tumors of the sigmoid colon declined by more than 2% per year. As in males, females demonstrated decreases of 3%–4% for in situ tumors at every subsite.

As observed in NH white females, Asian/PI females showed decreases in the incidence of tumors of the rectum and sigmoid colon, but increases in localized tumors of the proximal colon. Unlike NH white females, Asian/PI females experienced an annual increase in late-stage proximal colon tumors of 1.8% (nonsignificant). In situ tumors of the rectum and proximal colon declined by more than 6% per year during this period among Asian/PI females.

Black females also demonstrated declines in the incidence of tumors of the sigmoid colon, but increases in localized tumors of the proximal colon. Unlike NH white and Asian/PI females, black females had a 5% annual decline of in situ tumors in the sigmoid colon, but only minor declines of in situ tumors in the rectum and proximal colon. Among black females, the incidence of late-stage rectal tumors increased by 1.6% per year (not significant).

Hispanic females, as did Hispanic males, experienced an increased incidence in localized tumors of the sigmoid and proximal colon. However, Hispanic females had a pronounced decrease (nearly 4% per year) in the incidence of regional and distant tumors of the sigmoid colon. As other racial/ethnic groups did, Hispanic females had a decline of in situ tumors at every subsite. The incidence of late-stage tumors of the rectum and proximal colon remained unchanged among Hispanic females.

Incidence of Colorectal Cancer by Gender and SEER Registry

The population in the Alaska Natives registry was not large enough to allow calculation of EAPCs by stage of diagnosis and tumor subsite. Therefore, the results for 9 SEER registries, and the San Francisco-Oakland and San Jose registries combined, are presented in Tables 4 and 5.

Table 4. Estimated Annual Percent Change in the Age-Adjusted Incidence Rates of Cancer of the Colon and Rectum, by Subsite, Stage at Diagnosis, and Registry: Males, SEER* Program, 1992–2001
SubsiteStageAtlantaConnecticutDetroitHawaiiIowaLos AngelesNew MexicoSan Francisco Bay AreaSeattle/Puget SoundUtah
  • *

    SEER registries cover ˜14% of the U.S. population.

  • “All stages” category includes unstaged cases.

  • All sites” category includes appendix and large intestine, not otherwise specified.

  • §

    The estimated annual percent change is significantly different from zero (P < 0.05).

RectumIn situ−0.13.5−2.20.4−1.3−3.1−2.6−2.9−12.5§1.0
Localized−0.62.0§0.31.32.1−0.3−0.2−0.41.02.3
Regional/Distant0.7−1.0−1.60.8−1.0−1.10.8−2.2§−0.3−1.5
All stages−1.00.4−1.00.80.2−1.2§0.3−2.1§−1.00.1
Sigmoid colonIn situ−2.7−1.6−1.4−2.9−7.3§−2.8−4.6−4.5−10.7§2.6
Localized−6.4§−0.6−1.5§−3.0−0.7−2.3−5.6§−2.6§−2.7§−1.3
Regional/Distant−2.8−3.1§−0.7−2.0−2.9−0.1−4.0§−2.2−3.4§−1.1
All stages−3.8§−2.0§−1.5−2.7§−2.4§−1.2−4.6§−2.7§−3.5§−1.6
Proximal colonIn situ−0.6−3.8−1.2−1.7−0.5−3.3§−1.7−5.6−7.33.0
Localized0.7−0.6−0.10.42.8§−1.12.7−0.51.6§0.2
Regional/Distant−1.5−0.8−1.11.5−0.4−1.5§−0.3−2.2§−0.1−1.2
All stages−0.4−1.1−1.10.60.6−1.4§0.7−2.0§0.0−0.5
All sitesIn situ−0.4−0.5−1.4−1.8−3.1§−3.1§−2.5−4.0−10.7§0.4
Localized−1.30.4−0.3−0.51.6§−1.1§−0.3−1.00.10.6
Regional/Distant−0.8−1.5§−1.20.1−1.1−0.9−0.6−2.1§−0.9−1.4
All stages−1.0−0.9−1.2§−0.5−0.3−1.2§−0.7−2.2§−1.3§−0.8
Table 5. Estimated Annual Percentage Changes in the Age-Adjusted Incidence Rates of Cancer of the Colon and Rectum, by Subsite, Stage at Diagnosis, and Registry: Females, SEER* Program, 1992–2001
SubsiteStageAtlantaConnecticutDetroitHawaiiIowaLos AngelesNew MexicoSan Francisco Bay AreaSeattle/Puget SoundUtah
  • *

    SEER registries cover ˜14% of the U.S. population.

  • “All stages” category includes unstaged cases.

  • “All sites” category includes appendix and large intestine, not otherwise specified.

  • §

    The estimated annual percentage change is significantly different from zero (P < 0.05).

RectumIn situ−5.6−4.1−5.2−1.9−4.0−1.6−3.9−3.9−10.4§−2.9
Localized−0.81.41.31.40.90.7−0.8−1.3−1.91.5
Regional/Distant−0.61.50.10.1−1.7§1.1−1.8−4.5§0.30.3
All stages−1.90.50.5−0.1−0.60.4−1.4−3.4§−1.80.3
Sigmoid colonIn situ−8.0−2.3−2.8−0.1−6.1§−3.3−5.1−0.6−6.71.3
Localized−3.3§−0.41.0−2.1−0.6−2.0§−0.1−2.0§−1.4−0.3
Regional/Distant−5.2§−1.5−2.3§−2.2−3.0§−0.9−4.0−1.3−3.3§−3.2
All stages−4.6§−1.2−0.9−2.0−2.1§−1.5§−2.4−1.8§−2.8§−1.9
Proximal colonIn situ−3.81.9−4.6−8.1−2.3−1.3−0.9−8.3§−12.2§4.7
Localized3.10.32.1§−1.42.4§1.01.10.41.85.5§
Regional/Distant−3.1§0.9−1.20.1−0.2−0.3−0.6−1.0§0.5−1.3
All stages−1.30.5−0.3−1.50.50.10.1−1.0§0.51.4
All sitesIn situ−5.3§−0.9−4.5−3.9−4.5−2.1§−3.7−4.3§−9.0§−0.6
Localized0.20.51.6§−0.81.3§0.20.3−0.80.02.6
Regional/Distant−2.5§0.6−0.9§−0.6−1.1§0.2−1.4−1.7§−0.4−1.3
All stages−2.0§0.0−0.2−1.4−0.4−0.1−0.7−1.9§−0.90.1
Males

An overall decline in colorectal cancer among males was observed in all 10 SEER registries, for all sites and stages combined, although the decrease within individual SEER registries was not always statistically significant. Overall trends by registry are shown in Table 1, and trends by gender, subsite, and stage are shown in Table 4. The incidence of tumors in the sigmoid colon at any stage at diagnosis declined in all registries (except for a nonsignificant average annual increase of 2.6% for in situ tumors in Utah). Trends for rectal tumors had a less clear pattern across registries, varying from a significant decrease of 2.1% per year in the San Francisco Bay Area to a nonsignificant annual increase of 0.8% in Hawaii. The incidence of tumors of the proximal colon varied substantially by registry and stage at diagnosis. About half of the registries reported an increase in localized tumors, and the other half reported slight decreases. The incidence of in situ and late-stage tumors of the proximal colon declined in all registries, except for in situ tumors in Utah and regional or distant tumors in Hawaii.

Females

Among females, regional rates for all sites and stages of colorectal cancer combined did not decline as consistently as the rates among males (Table 1). Regardless of statistical significance, an overall decline was found in the incidence of tumors of the sigmoid colon at any stage at diagnosis (Table 5). The 2 exceptions to this finding were nonsignificant annual increases of 1.3% in Utah (in situ tumors) and 1.0% in metropolitan Detroit (localized tumors). The pattern for tumors of the rectum varied by registry and stage; about half of the registries reported increases and half reported decreases. The largest estimated changes for rectal tumors were decreases in the incidence of in situ tumors. The decrease in the incidence of in situ tumors of the proximal colon was observed in all but the Connecticut and Utah registries. Proximal colon tumors diagnosed at a localized stage appeared to have mostly increased during the period, whereas the incidence of late-stage tumors varied from a significant decrease of 3.1% per year in Atlanta to a nonsignificant annual increase of 0.9% in Connecticut.

Reporting Delay

To assess how much of the observed decline in colorectal cancer rates between 1992 and 2001 might be due to reporting delays, we recalculated rates and EAPCs by gender, stage at diagnosis, and subsite by using data from the November 2004 SEER data submission. Changes in EAPC values ranged from 0 to −0.2, for an average of 0.125 (data not shown). As an example, the original EAPCs for all sites and all stages combined (−1.2% for males and −0.7% for females) changed to −1.1% and −0.6% for males and females, respectively.

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

These results document continued declines in the incidence of cancers of the colon and rectum that have been observed since the 1950s.2 Rates of colorectal cancer decreased ˜1% per year between 1992 and 2001, and the decreases were most pronounced among males, among NH whites, and for tumors of the sigmoid colon. Delays in the reporting of cancer cases can result in underestimation of the true rates;13 however, repeat analysis using more recent data did not show a substantial change in the estimates. It does not appear that the continued decline in the incidence of these cancers can be attributed to reporting delay.

An evaluation of 1975–1990 data from the SEER program showed a decline in the rates of distant disease since 1975. Rates of regional disease increased until the early 1980s, followed by a decline in the late 1980s. Similarly, rates of local disease increased until the mid–1980s and then declined in the late 1980s. Rates of in situ disease increased until the late 1980s, followed by a recent decline.2 The current analysis documents a dramatic decrease of in situ tumors at all subsites between 1992 and 2001. These patterns indicate a stage shift consistent with increasingly earlier detection of colorectal cancer over time, probably as a result of the detection and removal of polyps.

The results by subsite also suggest an effect of screening. An evaluation of SEER data by subsite through 1994 demonstrated a decrease in the incidence of invasive colorectal cancer that was most evident in the distal (descending and sigmoid) colon and rectum.14 In the current study, declines were most dramatic for tumors of the sigmoid colon, and these changes may be attributable to the increased use of sigmoidoscopy for screening. Screening for cancer of the colon and rectum may result in both a shift to an earlier stage and a decrease in the incidence of all stages. Colonoscopy actually decreases the incidence of tumors by the removal of premalignant lesions.15 Most screening guidelines recommend fecal occult blood test, sigmoidoscopy, or both for men and women older than 50 years of age.16, 17 The flexible sigmoidoscope routinely reaches only the rectum and sigmoid colon,11 and the effect of screening with sigmoidoscopy may be confined primarily to the incidence of tumors at these subsites. An effect of screening also was confirmed by our analysis of trends by age group. A decline in colorectal cancer rates was seen for males and females older than 50 years of age, whereas little or no decline was observed among those younger than 50 years, for whom screening is not recommended.

The National Health Interview Surveys (NHIS) evaluated the use of colorectal cancer screening (fecal occult blood test within 2 years or endoscopy within the past 3 years).3 The results showed that the proportion of adults older than 50 years reporting recent use of colorectal cancer screening increased between 1987 and 1998, from 24% to 30% among females and from 22% to 37% among males.3 The proportion reporting recent screening in the 2000 survey was 41% for males and 37.5% for females.18 The greater increase for males than for females may account for the greater decline in incidence among males in our study.

Our analysis by race/ethnicity and SEER registry demonstrated that not all groups experienced a decline in cancer incidence, probably because of variation in the utilization of colorectal cancer screening. NH white males and females had the greatest decrease in cancer incidence. Blacks had only a small decline in incidence, although the increase in proximal colon cancer incidence demonstrated in the earlier SEER study14 appears to have leveled off in recent years. NHIS data showed that white males reported higher use of screening in 1998 (39%) than did black and Hispanic males (32% and 24%, respectively); similar results were observed for white, black, and Hispanic females (32%, 26%, and 18%, respectively).3 Results for Asians were not presented in the earlier NHIS report, but in the 2000 survey, Asians reported lower utilization of screening than did blacks and whites, but higher rates of screening than did Hispanics.18 This discrepancy is probably a result of differences in income, education, and insurance coverage, but cultural differences may also play a role. A study of screening practices among Latino and Anglo males and females enrolled in the Kaiser program in California reported lower use of screening tests by Latinos, even without financial barriers.19 Incidence rates of colorectal cancer actually increased slightly among California Hispanics during this period, but not as significantly as among New Mexico Hispanics between 1969 and 1994.20 Increased incidence is probably a result of changes in risk factors in this traditionally low-incidence group.20

We were unable to examine trends for Asian subgroups, and the Asian category combines several groups including Japanese, Chinese, Filipinos, and Hawaiians. Cancer incidence, screening practices, and risk factors for these groups are likely to vary considerably. A study of trends in colorectal cancer in California found that the incidence of rectal cancer declined between 1988 and 2001 among Chinese males, but not among other Asian subgroups.21 During the same timeperiod, the incidence of colon cancer increased among Korean males and females, probably because of the adoption of a western diet and lifestyle.21

Decreases in cancer rates also varied by SEER region of residence. Males residing in San Francisco, Los Angeles, and Seattle and females residing in Atlanta and San Francisco experienced the largest declines in cancer incidence. Analysis of data by metropolitan area from the Behavioral Risk Factor Surveillance System (BRFSS) showed that the use of sigmoidoscopy or colonoscopy in 2002 varied by state and metropolitan area; the highest prevalence was in Minnesota and the lowest prevalence was in Hawaii (males and females combined).22 SEER regions do not correspond with BRFSS areas, and no data were reported for the San Francisco Bay Area.

Despite the evidence of a screening effect, it is likely that other factors play a role in the decrease in rates of colorectal cancer. Although the most dramatic declines occurred in the sigmoid colon, some evidence indicated a decrease in late-stage and in situ tumors of the proximal colon. This decrease could be attributable to increased use of colonoscopy, but also might reflect a change in risk factors such as diet. Some studies have shown an association between red meat consumption and colon cancer,23, 24 and dietary behaviors including consumption of red meat are likely to vary by geographic region. The United States Department of Agriculture Food Consumption Surveys reported some improvements in healthful eating among blacks and whites between 1965 and 1991, including evidence of decreased consumption of fat and increased consumption of grains and legumes.4

A decreased incidence of colon cancer may also be the result of increased use of nonsteroidal anti-inflammatory drugs (NSAIDs). Use of aspirin and other NSAIDs has been shown to reduce the risk of colorectal cancer.25 A survey of physician visits showed an increase in NSAID prescriptions for both acute and chronic pain between 1980–1981 and 1999–2000, and the data showed a reduced threshold for prescribing NSAIDs for pain.26

The results of this analysis demonstrate that declines in the incidence of colon and rectal cancers observed in U.S. SEER areas during the 1970s and 1980s continued through the 1990s. It is likely that this decline is primarily due to the increased use of screening, especially sigmoidoscopy. Although a reduction in rates of colorectal cancer is welcome news, clearly more progress is needed. This disease remains the third most frequent cause of cancer death in the United States, and the prevalence of screening that could prevent it remains low. Further, some racial/ethnic groups are not experiencing equally the benefits of screening and the reductions in cancer incidence.

Acknowledgements

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

The collection of cancer incidence data used in this study was supported by the California Department of Health Services as part of the statewide cancer reporting program mandated by California Health and Safety Code Section 103885, the National Cancer Institute's Surveillance, Epidemiology, and End Results Program, and the Centers for Disease Control and Prevention National Program of Cancer Registries

We appreciate the in-kind support from all the contributors to this monograph and also are grateful to Faruque Ahmed for his leadership of the colorectal cancer monograph project

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES
  • 1
    Jemal A, Clegg LX, Ward E, et al. Annual report to the nation on the status of cancer, 1975–2001, with a special feature regarding survival. Cancer. 2004; 101: 327.
  • 2
    Chu K, Tarone R, Chow W, Hankey B, Ries L. Temporal patterns in colorectal cancer incidence, survival and mortality from 1950 through 1990. J Natl Cancer Inst. 1994; 86: 9971006.
  • 3
    Breen N, Wagener DK, Brown ML, Davis WW, Ballard-Barbash R. Progress in cancer screening over a decade: Results of cancer screening from the 1987, 1992, and 1998 national health interview surveys. J Natl Cancer Inst. 2001; 93: 17041713.
  • 4
    Popkin B, Siega-Riz A, Haines P. A comparison of dietary trends among racial and socioeconomic groups in the United States. N Engl J Med. 1996; 335: 716720.
  • 5
    Jackson-Thompson J, German R, Ahmed F, Lai S, Friedman C. Descriptive epidemiology of colorectal cancer in the United States, 1998–2001. Cancer. 107, ( 5 Suppl).
  • 6
    Surveillance, Epidemiology, and End Results (SEER) Program SEER*Stat Database: Incidence—SEER 11 Regs + AK Public-Use, Nov 2003 Sub for Expanded Races (1992–2001), National Cancer Institute, DCCPS, Surveillance Research Program, Cancer Statistics Branch, released April 2004, based on the November 2003 submission. Available at URL: www.seer.cancer.gov
  • 7
    Surveillance, Epidemiology, and End Results (SEER) Program SEER*Stat Database: Incidence—SEER 11 Regs Public-Use, Nov 2003 Sub for Hispanics (1992–2001), National Cancer Institute, DCCPS, Surveillance Research Program, Cancer Statistics Branch, released April 2004, based on the November 2003 submission. Available at URL: www.seer.cancer.gov
  • 8
    YoungJ, RoffersS, RiesL, FritzA, HurlbutAE, eds. SEER Summary Staging Manual, 2000. Codes and Coding Instructions. Bethesda, MD: National Cancer Institute; 2001.
  • 9
    FritzA, PercyC, JackA, et al., eds. International Classification of Diseases for Oncology. 3rd edn. Geneva: World Health Organization; 2000.
  • 10
    PercyC, Van HoltenV, MuirC, eds. International Classification of Diseases for Oncology. 2nd edn. Geneva: World Health Organization; 1990.
  • 11
    Painter J, Saunders DB, Bell GD, Williams CB, Pitt R, Bladen J. Depth of insertion at flexible sigmoidoscopy: Implications for colorectal cancer screening and instrument design. Endoscopy. 1999; 31: 227231.
  • 12
    Kleinbaum D, Kupper L, Muller K. Applied Regression Analysis and Other Multivariable Methods. North Scituate, MA: Duxbury Press; 1988.
  • 13
    Clegg LX, Feuer EJ, Midthune DN, Fay MP, Hankey BF. Impact of reporting delay and reporting error on cancer incidence rates and trends. J Natl Cancer Inst. 2002; 94: 15371545.
  • 14
    Troisi R, Freedman A, Devesa S. Incidence of colorectal carcinoma in the U.S.: An update of trends by gender, race, age, subsite, and stage, 1975–1994. Cancer. 1999; 85: 16701676.
  • 15
    Winawer S, Zauber A, Ho MN, et al. Prevention of colorectal cancer by colonoscopic polypectomy. N Engl J Med. 1993; 329: 19771981.
  • 16
    Pignone M, Rich M, Teutsch SM, Berg AO, Lohr KN. Screening for colorectal cancer in adults at average risk: Asummary of the evidence for the U.S. preventive services task force. Ann Intern Med. 2002; 137: 132141.
  • 17
    Smith RA, Cokkinides V, von Eschenbach AC, et al. American Cancer Society guidelines for the early detection of cancer. CA Cancer J Clin. 2002; 52: 822.
  • 18
    Swan J, Breen N, Coates RJ, Rimer BK, Lee NC. Progress in cancer screening practices in the United States: Results from the 2000 national health interview survey. Cancer. 2003; 97: 15281540.
  • 19
    Perez-Stable EJ, Otero-Sabogal R, Sabogal F, McPhee SJ, Hiatt RA. Self-reported use of cancer screening tests among Latinos and Anglos in a prepaid health plan. Arch Intern Med. 1994; 154: 10731081.
  • 20
    Chao A, Gilliland F, Hunt W, Bulterys M, Becker T, Key C. Increasing incidence of colon and rectal cancer among Hispanics and American Indians in New Mexico (United States), 1969–94. Cancer Causes Control. 1998; 9: 137144.
  • 21
    Cress R. Colon and rectum. In: CockburnM, DeapenD, eds. Cancer Incidence and Mortality in California; Trends by Race/Ethnicity, 1988–2001. Los Angeles: Los Angeles Cancer Surveillance Program, University of Southern California; 2004: 5862.
  • 22
    Balluz L, Ahluwalia IB, Murphy W, Mokdad A, Giles W, Harris VB. Surveillance for certain health behaviors among selected local areas—United States, behavioral risk factor surveillance system 2002. MMWR Surveill Summ. 2004; 53: 1100.
  • 23
    Le Marchand L, Wilkens L, Hankin J, Kolonel L, Lyu L. A case-control study of diet and colorectal cancer in a multiethnic population in Hawaii (United States): Lipids and foods of animal origin. Cancer Causes Control. 1997; 8: 637648.
  • 24
    Larsson SC, Rafter J, Holmberg L, Bergkvist L, Wolk A. Red meat consumption and risk of cancers of the proximal colon, distal colon and rectum: The Swedish mammography cohort. Int J Cancer. 2005; 113: 829834.
  • 25
    Smalley W, Ray W, Daugherty J, Griffin M. Use of nonsteroidal anti-inflammatory drugs and incidence of colorectal cancer: A population-based study. Arch Intern Med. 1999; 159: 161166.
  • 26
    Caudill-Slosberg MA, Schwartz LM, Woloshin S. Office visits and analgesic prescriptions for musculoskeletal pain in US: 1980 vs. 2000. Pain. 2004; 109: 514519.