Colorectal cancer statistics, 2023

Abstract

Colorectal cancer (CRC) is the second most common cause of cancer death in the United States. Every 3 years, the American Cancer Society provides an update of CRC statistics based on incidence from population-based cancer registries and mortality from the National Center for Health Statistics. In 2023, approximately 153,020 individuals will be diagnosed with CRC and 52,550 will die from the disease, including 19,550 cases and 3750 deaths in individuals younger than 50 years. The decline in CRC incidence slowed from 3%–4% annually during the 2000s to 1% annually during 2011–2019, driven partly by an increase in individuals younger than 55 years of 1%–2% annually since the mid-1990s. Consequently, the proportion of cases among those younger than 55 years increased from 11% in 1995 to 20% in 2019. Incidence since circa 2010 increased in those younger than 65 years for regional-stage disease by about 2%–3% annually and for distant-stage disease by 0.5%–3% annually, reversing the overall shift to earlier stage diagnosis that occurred during 1995 through 2005. For example, 60% of all new cases were advanced in 2019 versus 52% in the mid-2000s and 57% in 1995, before widespread screening. There is also a shift to left-sided tumors, with the proportion of rectal cancer increasing from 27% in 1995 to 31% in 2019. CRC mortality declined by 2% annually from 2011–2020 overall but increased by 0.5%–3% annually in individuals younger than 50 years and in Native Americans younger than 65 years. In summary, despite continued overall declines, CRC is rapidly shifting to diagnosis at a younger age, at a more advanced stage, and in the left colon/rectum. Progress against CRC could be accelerated by uncovering the etiology of rising incidence in generations born since 1950 and increasing access to high-quality screening and treatment among all populations, especially Native Americans.

INTRODUCTION

Colorectal cancer (CRC) is the third most commonly diagnosed cancer and the third most common cause of cancer-related death in both men and women in the United States. However, it ranks second in cancer-related deaths overall and is the leading cause in men younger than 50 years. More than one half of all cases and deaths are attributable to modifiable risk factors, such as smoking, an unhealthy diet, high alcohol consumption, physical inactivity, and excess body weight.¹ In addition, a large proportion of CRC incidence and mortality is preventable through the receipt of regular screening, surveillance, and high-quality treatment.² In this article, we provide a comprehensive overview of current CRC statistics in the United States, including the estimated numbers of new cases and deaths in 2023 by age and incidence, survival, and mortality rates and trends by age, race, and ethnicity based on incidence data through 2019 and mortality data through 2020. CRC screening prevalence for adults aged 45 years and older is also presented nationally for 2021 and by state for 2020.

MATERIALS AND METHODS

Data sources

Population-based cancer incidence data in the United States are collected by the National Cancer Institute's (NCI's) Surveillance, Epidemiology, and End Results (SEER) program and the Centers for Disease Control and Prevention's (CDC's) National Program of Cancer Registries. Combined SEER and National Program of Cancer Registries data for 1995 through 2019, as provided by the North American Association of Central Cancer Registries, are the source for national incidence trends; the estimated new CRC diagnoses in 2023; case distributions by stage, age, and subsite; and 5-year average annual incidence rates.³ Temporal trends in incidence rates were based on all states with available data during 1998–2019, covering 90% of the US population, and were adjusted for delays in case reporting based on national North American Association of Central Cancer Registries delay factors. Delay adjustment accounts for the additional time required for the complete registration of cases and more accurately reflects cancer trends in the most recent time period.⁴ Racial misclassification for American Indian and Alaska Native (AIAN) persons has been reduced by restricting incidence rates to Purchased/Referred Care Delivery Area counties.

Historical incidence trends dating back to 1975 are based on data from the eight oldest SEER registries (Connecticut, Iowa, Hawaii, New Mexico, Utah, and the metropolitan areas of Atlanta, San Francisco–Oakland, and Seattle–Puget Sound), representing approximately 8% of the US population.⁵ The SEER 17 catchment area (SEER 8 plus registries for Alaska Natives, rural and greater Georgia, San Jose–Monterey and greater California, Kentucky, Louisiana, and New Jersey) was the source for 5-year relative survival,⁶ and the complete SEER registries (SEER 17 plus Idaho, Illinois, Massachusetts, New York, and Texas), which achieve 48% population coverage, were the source for incidence rates by single year of age (Figure 1) and the lifetime probability of developing CRC.

FIGURE 1

Age-specific colorectal cancer incidence rates, 2015–2019, United States. Incidence rates exclude appendiceal cancer and are age-adjusted to the 2000 US standard population. Source: Main figure: North American Association of Central Cancer Registries, 2022; Inset: Surveillance, Epidemiology, and End Results Program, 2022.

US mortality data from 1930 to 2020 were obtained from the CDC's National Center for Health Statistics (NCHS).⁷ Detailed information on decedent race and ethnicity is limited to deaths occurring from 1990 onward. Mortality rates for AIAN persons are based on the entire US population and were adjusted for racial misclassification using classification factors previously published by the NCHS.⁸ Incidence and mortality rates for persons who are Alaska Native separate from American Indian were based on cases collected by the SEER program's Alaska Native Tumor Registry and deaths occurring in AIAN persons in Alaska as reported by the NCHS. Because of data limitations, there may be some cross-contamination in data presented for Alaska Native and American Indian persons separately.

CRC screening prevalence at the state level was obtained from the 2020 Behavioral Risk Factor Surveillance System (BRFSS) public-use data.⁹ The BRFSS was designed to provide state prevalence estimates for health behaviors and is coordinated by the CDC and conducted by individual state health departments. Data are collected from computer-assisted telephone interviews with adults aged 18 years and older. In 2011, the CDC modified the BRFSS weighting procedures and expanded to include households without landline telephone service (i.e., cellular service only).¹⁰ Therefore, BRFSS estimates for 2011 and later, including those herein, should not be compared with earlier estimates.

National CRC screening prevalence (2021) was obtained from the NCHS’ National Health Interview Survey (NHIS).¹¹ The NHIS is conducted by the US Census Bureau and was designed to provide national prevalence estimates on health behaviors, such as cancer screening. Data are collected through computer-assisted, in-person interviews of adults aged 18 years and older.

Projected new cases and deaths in 2023

The most recent year for which incidence and mortality data are available lags 2–4 years behind the current year because of the time required for data collection, compilation, quality control, and dissemination. Therefore, the American Cancer Society projects the numbers of new cancer cases and deaths in the United States in the current year to estimate the contemporary cancer burden. These estimates cannot be used for tracking cancer occurrence over time because they are model-based, and the methodology changes every few years, most recently in 2021, to incorporate improvements in statistical methods, increased cancer registration coverage, and covariate information. The methods for projecting the number of new CRC cases and deaths that will occur in 2023 overall and by age are described in detail elsewhere.^{12, 13}

Statistical analysis

CRC cases were classified according to the International Classification of Diseases for Oncology as colon (C18.0–C18.9 and C26.0) or rectum (C19.9 and C20.9).¹⁴ Colon tumors were further designated by anatomic location as proximal (C18.0 and C18.2–C18.4), distal (C18.5–C18.7), or not otherwise specified (C18.8, C18.9, and C26.0). All incidence rates exclude appendix (C18.1). Because of the large number of rectal cancer deaths that are misclassified as colon in death certificates,¹⁵ colon and rectal cancer deaths were combined in all analyses. This misclassification does not affect the calculation of relative survival rates, which is based on tumor subsite information in cancer registry data. All statistics presented herein by race, including those for Asian American/Pacific Islander (AAPI) and AIAN individuals, are exclusive of Hispanic ethnicity for improved accuracy of classification.

NCI's SEER*Stat program (version 8.4.0) was used to calculate age-adjusted (2000 US standard population using 19 age groups) CRC incidence and mortality rates, expressed per 100,000, and rate ratios (RRs) with accompanying 95% confidence intervals (CIs).¹⁶ Incidence and mortality trends were quantified using NCI's Joinpoint regression program (version 4.9.1.0).¹⁷ Trends were described as increasing or decreasing when the annual percent change was statistically significant based on a two-sided p value < .05 and otherwise were described as stable. The lifetime probability of developing cancer was obtained from the NCI's DevCan software program (version 6.8.0).¹⁸

SELECTED FINDINGS

Estimated cases and deaths in 2023

There will be an estimated 153,020 new cases of CRC in the United States in 2023, including 106,970 tumors in the colon and 46,050 tumors in the rectum. Although the majority of diagnoses occur in people 65 years and older, 19,550 cases (13%) will be in individuals younger than 50 years and one-third will be in individuals 50–64 years (Table 1). Approximately 43% of diagnoses before age 50 years, often referred to as early onset disease, are in people aged 45–49 years, who are now recommended to be screened. In addition, in 2023, there will be an estimated 52,550 CRC deaths, including 3750 decedents (7%) younger than 50 years.

TABLE 1. Estimated numbers of new invasive colorectal cancer cases and deaths by age, 2023, United States.

Age, years	Cases												Deaths^a
	Colon & rectum				Colon				Rectum				Colorectum
	Male	Female	Total	Percent	Male	Female	Total	Percent	Male	Female	Total	Percent	Male	Female	Total	Percent
Birth to 49	10,560	8990	19,550	13%	6410	5960	12,370	12%	4150	3030	7180	16%	2150	1,600	3750	7%
50–64	28,810	19,400	48,210	32%	17,500	12,300	29,800	28%	11,310	7100	18,410	40%	8030	5130	13,160	25%
≥65	42,490	42,770	85,260	56%	30,510	34,290	64,800	61%	11,980	8480	20,460	44%	18,290	17,350	35,640	68%
All ages	81,860	71,160	153,020	100%	54,420	52,550	106,970	100%	27,440	18,610	46,050	100%	28,470	24,080	52,550	100%

• Note: Estimates are rounded to the nearest 10 and exclude in situ carcinoma.
• ^a Colon and rectal cancer deaths are not presented separately because of the large rate of misclassification.

Incidence

The risk of CRC escalates rapidly with age; during 2015–2019, incidence rates increased by 80%–100% with each 5-year age group until age 50 years and then by 20%–30% from ages 55–59 years and older (Figure 1). However, there is only a 9% increase from ages 50–54 to 55–59 years (from 60.6 to 66.1 cases per 100,000 population per year) because the natural age-related pattern of detecting symptomatic CRC is disrupted by the detection of precancerous adenomas and prevalent cancers in asymptomatic persons through the introduction of screening, which was recommended to begin at age 50 years until 2018.¹⁹ Notably, the incidence rate for individuals who are aged 50 years (71.2 per 100,000) is 4%–14% higher than that at ages 51–58 years, despite just 46% screening prevalence in those aged 50–54 years in 2019.²⁰ The inset of Figure 1 shows age-specific rates in the absence of screening during the mid-1970s for comparison (red bars).

The average annual overall incidence rate during 2015 through 2019 was 33% higher in men (41.5 per 100,000) than in women (31.2 per 100,000; Figure 2). However, the sex disparity varies by age at diagnosis and tumor location. For example, the male-to-female incidence RR ranges from 1.10 (95% CI, 1.09–1.11) for proximal colon cancer to 1.61 (95% CI, 1.60–1.63) for rectal cancer, as reflected in the relatively large differences in tumor distribution (Table 2). Similar sex differences have been reported in other high-income countries²¹ and likely reflect variation in environmental and endogenous carcinogenic and hormonal exposures combined with distinct etiologic mechanisms across anatomic subsites.^22-26 A recent study estimated that approximately one third of the CRC sex disparity is attributed to differences in the prevalence of established risk factors.²⁷

FIGURE 2

Colorectal cancer incidence (2015–2019) and mortality (2016–2020) rates by sex, race, and ethnicity, United States. Incidence rates exclude appendiceal cancer. ^aTo reduce racial misclassification, incidence rates are limited to Purchased/Referred Care Delivery Area counties, and mortality rates (for the entire United States) are adjusted using factors published by the National Center for Health Statistics.⁸ Incidence and mortality rates are age-adjusted to the 2000 US standard population. AI, American Indian; AN, Alaska Native; AAPI, Asian American/Pacific Islander; NHB, non-Hispanic Black; NHW, non-Hispanic White. Source: Incidence: North American Association of Central Cancer Registries, 2022; Mortality: National Center for Health Statistics, 2022. [Correction added on 30 March 2023, after first online publication: Figure 2 has been revised.]

TABLE 2. Colorectal cancer incidence rates and proportions by tumor subsite, sex, and age, United States, 2015 to 2019.

	Overall		Male		Female		20–49 years		50–64 years		>65 years
	Rate	Proportion	Rate	Proportion	Rate	Proportion	Rate	Proportion	Rate	Proportion	Rate	Proportion
Proximal	13.9	39%	14.6	35%	13.3	44%	2.8	23%	19.6	29%	77.2	47%
Distal	8.8	24%	10.6	26%	7.4	23%	3.7	30%	19.4	28%	34.1	21%
Rectal	11.0	30%	13.9	34%	8.6	27%	5.1	42%	26.7	38%	37.9	24%
Large Intestine NOS	2.1	6%	2.4	6%	1.9	6%	0.5	4%	3	4%	11.7	7%
Colorectum	35.9	100%	41.5	100%	31.2	100%	12	100%	69	100%	160.9	100%

• Note: Excludes appendiceal cancer.
• Abbreviation: NOS, not otherwise specified.

Trends

The overall annual, age-standardized CRC incidence rate has decreased by 46%, from 66.2 per 100,000 at its peak in 1985 to 35.7 per 100,000 persons in 2019. The decline is very similar in men and women (Figure 3) and is attributed to changing patterns in risk factors, such as reductions in smoking and increased use of nonsteroidal anti-inflammatory drugs, and the uptake of CRC screening among individuals aged 50 years and older.²⁸ In particular, the rapid dissemination of colonoscopy screening, which has a greater capacity for CRC prevention than other recommended tests, is credited with especially steep declines in incidence among adults aged 50 years and older—about 3%–5% annually in the late 2000s (Table 3)—after Medicare expanded coverage to all beneficiaries in 2001.^29-31 Colonoscopy prevalence tripled from 20% in 2000 to 61% in 2018 among adults aged 50 years and older.³² (See section on Colorectal cancer screening for more information).

FIGURE 3

Trends in colorectal cancer incidence (1975–2019) and mortality (1930–2020) rates by sex, United States. Because of changes in the International Classification of Diseases coding for mortality, numerator information has changed over time. Incidence rates exclude appendiceal cancer, are age-adjusted to the 2000 US standard population, and adjusted for reporting delays. Source: Incidence: Surveillance, Epidemiology, and End Results Program, 2022; Mortality: National Center for Health Statistics, 2022.

TABLE 3. Trends in colorectal cancer incidence rates by age and stage at diagnosis, 1998–2019, United States.

	Trend 1		Trend 2		Trend 3		Trend 4		Trend 5		2010–2019 AAPC	2015–2019 AAPC
	Years	APC	Years	APC	Years	APC	Years	APC	Years	APC
All stages
20–49 years	1998-2008	1.4^a	2008-2011	0.2	2011-2019	1.9^a					1.7^a	1.9^a
50–64 years	1998-2001	−0.6	2001-2007	−1.8^a	2007-2011	−3.1^a	2011-2014	1.7	2014-2019	−0.1	0.1	−0.1
≥65 years	1998-2002	−2.2^a	2002-2008	−3.7^a	2008-2012	−4.9^a	2012-2019	−3.0^a			−3.4^a	−3.0^a
All ages	1998-2002	−1.6^a	2002-2008	−2.8^a	2008-2011	−4.1^a	2011-2019	−1.4^a			−1.7^a	−1.4^a
Localized
20–49 years	1998-2007	2.9^a	2007-2019	−1.1^a							−1.1^a	−1.1^a
50–64 years	1998-2007	0.8^a	2007-2011	−5.1^a	2011-2014	1.2	2014-2019	−2.9^a			−1.8	−2.9^a
≥65 years	1998-2006	−0.4	2006-2019	−5.7^a							−5.7^a	−5.7^a
All ages	1998-2006	0.1	2006-2019	−4.3^a							−4.3^a	−4.3^a
Regional
20–49 years	1998-2010	0.2	2010-2019	3.3^a							3.3^a	3.3^a
50–64 years	1998-2002	−2.0^a	2002-2005	−6.0^a	2005-2011	−2.7^a	2011-2019	2.2^a			1.6^a	2.2^a
≥65 years	1998-2002	−3.0^a	2002-2005	−7.8^a	2005-2013	−4.1^a	2013-2019	−0.7			−1.8^a	−0.7
All ages	1998-2002	−2.5^a	2002-2005	−6.9^a	2005-2013	−3.0^a	2013-2016	2.4	2016-2019	−0.6	−0.4	0.1
Distant
20–49 years	1998-2019	2.7^a									2.7^a	2.7^a
50–64 years	1998-2000	−3.1	2000-2005	0.3	2005-2009	−2.0	2009-2019	0.5^a			0.5^a	0.5^a
≥65 years	1998-2000	−4.0^a	2000-2004	−1.0	2004-2017	−2.6^a	2017-2019	−0.3			−2.1^a	−1.5
All ages	1998-2000	−3.4	2000-2005	−0.3	2005-2010	−1.9^a	2010-2019	−0.5^a			−0.5^a	−0.5^a
Unknown stage
20–49 years	1998-2000	7.6	2000-2005	−5.1^a	2005-2013	−1.0	2013-2019	3.7^a			2.1^a	3.7^a
50–64 years	1998-2001	2.9	2001-2004	−11.0^a	2004-2012	−2.1^a	2012-2016	5.1^a	2016-2019	−0.4	1.6	0.9
≥65 years	1998-2001	−0.9	2001-2004	−9.2^a	2004-2012	−4.8^a	2012-2019	−1.0^a			−1.8^a	−1.0^a
All ages	1998-2001	−0.0	2001-2004	−9.4^a	2004-2012	−4.0^a	2012-2019	0.2			−0.7^a	0.2

• Note: Excludes appendiceal cancer. Trends based on age-standardized incidence rates adjusted for delays in case reporting and analyzed by the Joinpoint Regression Program, version 4.9.1.0, allowing up to four joinpoints.
• Abbreviations: APC, annual percent change; AAPC, average annual percent change.
• ^a The APC or AAPC is significantly different from zero (p < .05).
• Data source: North American Association of Central Cancer Registries, 2022.

The introduction of colonoscopy and widespread detection of prevalent, asymptomatic cancers resulted in a shift to earlier stage diagnosis. The proportion of localized stage disease increased from 33% in 1995 to 41% in 2005, with a parallel decline in regional stage disease from 39% to 34% (Figure 4). This is similar to the stage shift that emerged in the 1980s because of earlier detection through improved diagnostic techniques and increased use of fecal occult blood tests and sigmoidoscopy,³³ which were recommended by the American Cancer Society for CRC screening as early as 1977.^{34, 35} Over the past decade, however, this pattern has reversed as steep declines in the incidence of localized-stage disease (4.3% annually from 2006 to 2019) has coincided with an uptick in advanced diagnoses (Figure 5). Since circa 2010, the incidence of regional-stage and distant-stage disease has increased by about 3% per year in people younger than 50 years and by 2% and 0.5% per year, respectively, in people aged 50–64 years, while rates in people aged 65 years and older stabilized since about 2015 after a decade of steep decline (Table 3). Consequently, a crossover occurred between incident cases of localized-stage and regional-stage disease in 2015–2016 overall, among individuals aged 50 years and older (Figure 5), and for every anatomic subsite (proximal and distal colon and rectum). In 2019, 60% of new cases were advanced disease (including 22% with distant metastases), up from 52% during 2005–2008 and 57% in 1995, before widespread screening. The shift to later stage diagnosis is only partly explained by improved staging, as the incidence of unstaged disease declined from 2001 to 2012 but remained stable thereafter (Table 3). Other contributing factors likely include screening saturation (i.e., diminished proportion of initial uptake)³⁶ and its disproportionate detection and removal of slow-growing (vs. more aggressive) adenomas,^37-39 as well as the increasing influence of earlier onset cancers, which are more often advanced.⁴⁰

FIGURE 4

Trends in colorectal cancer stage distribution (%), 1995–2019, United States. Excludes appendiceal cancer. Source: North American Association of Central Cancer Registries, 2022.

FIGURE 5

Trends in the incidence of colorectal cancer by age and stage at diagnosis, 1998–2019, United States. Incidence rates exclude appendiceal cancer, are age-adjusted to the 2000 US standard population, and adjusted for reporting delays. Source: North American Association of Central Cancer Registries, 2022.

A similar shift is occurring in tumor location. The decline in incidence tapered over the past decade for all anatomic subsites but was most notable for rectal cancer because of rising incidence in individuals younger than 65 years (by about 2% and 1% per year in those aged 50 years and younger and aged 50–64 years, respectively) and a slowing decline in those aged 65 years and older (Figure 6). Consequently, the overall incidence of rectal cancer was similar to that of distal colon cancer before 2005 but was approximately 30% higher in 2019. The proportion of CRCs occurring in the rectum has increased steadily from 27% in 1995 to 31% in 2019, despite greater efficacy of screening for preventing left-sided lesions.^{41, 42} Approximately 4 in 10 diagnoses in people aged 50–64 years are now rectal tumors (Table 2). This pattern suggests changes in the etiology as well as the magnitude of underlying disease risk that was foreshadowed by trends in early onset CRC.

FIGURE 6

Trends in the incidence of colorectal cancer by age and tumor subsite, 1998–2019, United States. NOS indicates not otherwise specified. Incidence rates exclude appendiceal cancer, are age-adjusted to the 2000 US standard population, and are adjusted for reporting delays. Source: North American Association of Central Cancer Registries, 2022.

Early onset

Cancer trends in young adults are the best indicator of progress against disease because they reflect the influence of contemporary exposures as opposed to the cumulative long-term exposures manifest in older adults.⁴³ In contrast to decreasing CRC incidence in older adults, rates have been increasing in adults aged 20–39 years since the mid-1980s and in those aged 40–54 years since the mid-1990s.⁴⁴ From 2011 through 2019, rates increased by 1.9% per year in people younger than 50 years and in those aged 50–54 years. Steep increases in young and middle-aged adults alongside decreasing trends in older adults has rapidly shifted the patient population younger. For example, the proportion of newly diagnosed individuals who were younger than 55 years has almost doubled, from 11% in 1995 to 20% in 2019, despite this age group shrinking in the overall population from 80% to 71% during that time; for rectal cancer, which is largely driving the early-onset trend, the proportion has increased from 15% to 28%. Over the most recent decade, incidence in individuals younger than 50 years increased by about 2% per year for rectal cancer versus 0.5% and 1.6% per year for tumors in the proximal and distal colon, respectively (Figure 6). Early onset CRC incidence rose in every racial and ethnic group in the United States from 2010 through 2019, with the annual percent change ranging from 0.2% (confined to rectal tumors) in Black individuals to 1% in AAPI individuals, 2% in non-Hispanic White individuals, and 3% in Hispanic and AIAN individuals.

Elevated risk in generations born since the 1950s, referred to as a birth cohort effect,⁴⁵ is carried forward as individuals age and is already apparent in trends for people aged 50–64 years, which more closely resemble younger than older age groups since circa 2010 (Figures 5 and 6). Incidence is uniquely increasing for early onset disease in many high-income countries^{46, 47} for reasons that remain unclear. Likely contributors include obesity,^48-50 recently shown to be associated similarly with all major molecular subtypes of sporadic disease,⁵¹ and reduced dietary quality.^52-57

Although people who harbor germline variations that increase CRC risk (e.g., mutations in DNA mismatch repair genes associated with Lynch syndrome) are more likely to develop disease early, most cases before age 50 years are sporadic. Research has shown that sporadic early onset CRC is biologically similar to later onset disease but with several clinical distinctions.^{58, 59} For example, patients younger than 50 years are more likely to be women, with the male-to-female incidence RR of 1.20 (95% CI, 1.18–1.22) versus 1.45 (95% CI, 1.44–1.46) in those aged 50–64 years and 1.31 (95% CI, 1.30–1.32) in those 65 years or older.⁶⁰ In addition, younger patients are more likely to present with hematochezia (e.g., 41% vs. 26%) and abdominal pain,^{40, 59} at least in part because of the predominance of left-sided tumors (e.g., 73% vs. 53%).³ Early onset patients are also more often diagnosed with advanced disease, including 27% with distant metastases versus 20% of older patients (Figure 7). Moreover, the rising trend is confined to advanced diagnoses; from 2010 to 2019, rates increased by about 3% per year for regional-stage and distant-stage disease versus a decline of 1% per year for localized-stage disease (Table 3). Even the incidence of unknown stage, most of which is metastatic, increased by 3.7% annually from 2013 to 2019 in contrast with the usual downward trend because of staging improvements.

FIGURE 7

Colorectal cancer stage distribution (%) by (A) age and (B) race/ethnicity, 2015–2019, United States. AI/AN indicates American Indian/Alaska Native; AAPI, Asian American/Pacific Islander; NHB, non-Hispanic Black; NHW, non-Hispanic White. Excludes appendiceal cancer. ^aLimited to Purchased/Referred Care Delivery Area counties. Source: North American Association of Central Cancer Registries, 2022.

The higher prevalence of advanced disease in younger versus older patients is only partly explained by screening. A study of symptomatic patients found a 40% longer time to diagnosis among individuals younger than 50 years versus older individuals, including both longer duration of symptoms and work-up time, often because of misdiagnosis with more common conditions.⁴⁰ Nevertheless, survival rates based on cancer registry data are higher for early onset versus later onset patients overall (68% vs. 64%) and by stage. Older individuals have more comorbidities and are also less likely than their younger counterparts to receive aggressive treatment, including surgery, multiagent systemic therapy, and adjuvant or neoadjuvant therapies.^61-64 However, aggressive disease management does not always beget better survival and is often associated with toxicity that negatively affects long-term quality of life. Overtreatment of early onset disease is an increasing concern and is the focus of ongoing clinical trials.^{65, 66} Notably, individuals aged 50–64 years have the highest overall survival (70%; Figure 8), likely reflecting more early detection (35% localized-stage disease vs. 26% in patients younger than 50 years; Figure 7). Studies suggest that tumor biology is a more important prognostic indicator for CRC than age.^{67, 68}

FIGURE 8

Colorectal cancer 5-year relative survival (%) by (A) age and (B) race/ethnicity, 2012–2018, United States. AAPI indicates Asian American/Pacific Islander; AI/AN, American Indian/Alaska Native; NHB, non-Hispanic Black; NHW, non-Hispanic White. Rates are based on cases diagnosed from 2012 to 2018, all followed through 2019, and exclude appendiceal cancer. ^aLimited to Purchased/Referred Care Delivery Area counties; statistic based on small case numbers, particularly for distant-stage disease. Source: Surveillance, Epidemiology, and End Results Program, 2022.

Survival

The 5-year relative survival rate for CRC increased from 50% in the mid-1970s to 65% during 2012–2018. Long-term gains reflect earlier detection through routine clinical examinations (e.g., digital rectal examination) and screening;^{33, 69} more accurate staging through advances in imaging (e.g., positron emission tomography);⁷⁰ improvements in infection control and surgical techniques; and advances in chemotherapy and radiation.⁶² Better survival accompanied by population aging and growth has resulted in more than 1.4 million CRC survivors in the United States as of January 1, 2022.⁷¹ Approximately 10% of survivors are living with metastatic cancer, 44% of whom were initially diagnosed with early-stage disease.⁷² Nevertheless, stage at diagnosis is the most important predictor of survival, with 5-year relative survival ranging from 91% for localized disease to 14% for distant disease (Figure 8).

Patients have begun living longer with advanced disease in recent years because of improved surgical methods and increased cancer-directed surgery,⁷³ advances in the treatment of liver metastases,^74-76 and the development and optimization of targeted therapies.^{62, 77-81} The largest survival gains are for metastatic rectal cancer, with 30% of patients diagnosed during 2016–2018 surviving 3 years compared with 25% only a decade earlier.⁶ However, not all individuals have benefited; survival among Black patients remained stagnant at 21%–22%, likely caused by unequal dissemination of these advances.⁸² Black patients are less likely to be represented in clinical trials and to receive the molecular tumor profiling necessary for the use of targeted therapies against the epidermal growth factor and programmed death 1 (PD-1) receptors.⁸³ Some studies indicate that Black patients with equal access to care have treatment and survival comparable to those of White patients,⁸⁴ whereas others do not,^85-87 highlighting the need for parity across the cancer continuum in addressing disparities.⁸⁸ Doubeni and colleagues recently reported the successful elimination of racial disparities in CRC mortality at a large health care system with the implementation of a continuum-of-care screening program that included equitable, timely delivery of treatment.⁸⁹

Mortality

Overall CRC death rates decreased by 57% between 1970 (29.2 per 100,000) and 2020 (12.6 per 100,000). Although contemporary trends are strikingly similar by sex, early declines were entirely driven by women, among whom rates were higher than those in men in the 1930s but began decreasing in the late 1940s, almost 4 decades earlier than in men (Figure 3). Given that survival rates were improving in both men and women during this time, these patterns likely reflect sex differences in incidence trends.³³

Similar to incidence, the decline in mortality has slowed from about 4% annually during the early 2000s to about 2% from 2012 through 2020 (Table 4). Since 2012, the overall decline in mortality has outpaced incidence (2% vs. 1.4%), partly because improvements in treatment have extended survival. Nevertheless, trends vary widely by age and race and ethnicity; among individuals 50–64 years the rate has declined by 0.5% annually since 2005 overall but has stabilized in White persons since 2014 and increased by 0.5% annually since 1990 in AIAN persons. Among individuals younger than 50 years, the death rate continued to increase steadily by 1% per year since 2004 overall, including increases of 1.7% annually in White persons and approximately 3% annually in Hispanic and AIAN persons; however, rates have decreased by 0.6% per year since 1990 in Black individuals (Table 4).

TABLE 4. Trends in colorectal cancer mortality rates by age, race, and ethnicity, 1990–2020, United States.

	Trend 1		Trend 2		Trend 3		Trend 4		2011–2020 AAPC	2016–2020 AAPC
	Years	APC	Years	APC	Years	APC	Years	APC
All races combined
20–49 years	1990-1998	−0.8	1998-2001	2.1	2001-2004	−2.6	2004-2020	1.2^a	1.2^a	1.2^a
50–64 years	1990-2002	−1.9^a	2002-2005	−4.4^a	2005-2020	−0.5^a			−0.5^a	−0.5^a
≥65 years	1990-2001	−1.7^a	2001-2011	−3.4^a	2011-2020	−2.9^a			−2.9^a	−2.9^a
All ages	1990-2002	−1.7^a	2002-2005	−3.8^a	2005-2012	−2.5^a	2012-2020	−1.9^a	−2.0^a	−1.9^a
Non-Hispanic White
20–49 years	1990-2004	−0.2	2004-2020	1.7^a					1.7^a	1.7^a
50–64 years	1990-2002	−2.1^a	2002-2005	−4.5^a	2005-2014	−0.6^a	2014-2020	0.4	0.0	0.4
≥65 years	1990-2001	−1.8^a	2001-2011	−3.5^a	2011-2020	−2.7^a			−2.7^a	−2.7^a
All ages	1990-2002	−1.8^a	2002-2005	−3.9^a	2005-2011	−2.6^a	2011-2020	−1.8^a	−1.8^a	−1.8^a
Non-Hispanic Black
20–49 years	1990-2020	−0.6^a							−0.6^a	−0.6^a
50–64 years	1990-1999	−0.7	1999-2020	−1.8^a					−1.8^a	−1.8^a
≥65 years	1990-2001	−0.8^a	2001-2020	−3.5^a					−3.5^a	−3.5^a
All ages	1990-2001	−0.6^a	2001-2020	−2.8^a					−2.8^a	−2.8^a
Asian/Pacific Islander
20–49 years	1990-2020	0.1							0.1	0.1
50–64 years	1990-2005	−2.2^a	2005-2020	0.2					0.2	0.2
≥65 years	1990-2010	−1.8^a	2010-2020	−3.4^a					−3.4^a	−3.4^a
All ages	1990-2020	−1.8^a							−1.8^a	−1.8^a
American Indian/Alaska Native
20–49 years	1990-2020	2.8^a							2.8^a	2.8^a
50–64 years	1990-2020	0.5^a							0.5^a	0.5^a
≥65 years	1990-2013	−0.3	2013-2020	−5.0^a					−4.0^a	−5.0^a
All ages	1990-2011	0.3	2011-2020	−2.5^a					−2.5^a	−2.5^a
Hispanic
20−49 years	1990-2011	−0.1	2011-2020	3.1^a					3.1^a	3.1^a
50–64 years	1990-2020	−0.7^a							−0.7^a	−0.7^a
≥65 years	1990-2002	0.2	2002-2020	−2.3^a					−2.3^a	−2.3^a
All ages	1990-2001	0.1	2001-2020	−1.7^a					−1.7^a	−1.7^a

• Note: Trends based on age-standardized mortality rates analyzed using the Joinpoint Regression Program, version 4.9.1.0, allowing up to five joinpoints.
• Abbreviations: APC, annual percent change; AAPC, average annual percent change.
• ^a The APC or AAPC is significantly different from zero (p < .05).
• Source: National Center for Health Statistics, 2022.

Despite similar trends, the overall death rate is currently 43% higher in men (15.7 per 100,000) than in women (11 per 100,000; Figure 2), which is larger than the gap for incidence (33% higher). Men have a slightly lower 5-year relative survival rate (64%) compared with women (65%) despite a more favorable tumor subsite distribution. For example, 35% of men versus 44% of women develop tumors in the proximal (right) colon (Table 2), which have a higher risk of death compared with left-sided cancers independent of histological and molecular characteristics.^{90, 91} However, the largest sex disparity in 5-year survival is also for left-sided tumors at 66% in men versus 68% in women for distal colon cancer and 67% versus 70%, respectively, for rectal cancer. Sex disparities in CRC risk and outcomes are thought to be influenced by differences in both environmental and endogenous exposures.⁹² Estrogen appears to confer a protective influence in tumor development and progression, although the mechanism remains unknown.^{93, 94} There are also sex differences in tumor molecular characteristics that likely influence response to treatment.⁹⁵

Racial/ethnic disparities

CRC incidence, survival, and mortality vary substantially by race and ethnicity. Among the five major groups depicted in Figure 2, incidence and mortality rates are highest in AIAN and non-Hispanic Black (hereafter, Black) individuals and lowest in AAPI individuals. Racial disparities are generally largest for mortality among men. For example, compared with White men, CRC death rates are 46% higher in AIAN men and 44% higher in Black men, versus incidence gaps of 37% and 21%, respectively. Over the past decade, the racial disparity in incidence for both sexes combined narrowed from 25% to 15% for Black individuals⁶⁰ but widened for AIAN individuals from 21% to 41%.⁹⁶

Importantly, aggregated rates for these broadly defined racial categories defined by the federal government mask cancer occurrence within these heterogeneous populations. A recent study reported that among the relatively low-risk Asian American population, Native Hawaiian and Pacific Islander men have 26% higher CRC death rates than White men.⁹⁷ Even more alarming is the disproportionate burden in Alaska Native individuals, among whom incidence rates are more than twice those in White individuals (88.5 vs. 35.7 per 100,000) and mortality rates are almost three times higher (35.9 vs. 13.1 per 100,000) [Correction added on 30 March 2023, after first online publication: In preceding sentence "...almost four times higher (50.5 vs..." has been changed to "...almost three times higher (35.9 vs..." in this version.]. CRC is the most commonly diagnosed cancer in Alaska Native persons, who have the highest rates in world.^{98, 99} In contrast to declining rates in all other racial and ethnic groups, mortality is increasing by more than 2% annually in Alaska Native persons.^{98, 100} Furthermore, among individuals younger than 50 years, rates are increasing by 4% per year for incidence and by 3% per year for mortality.¹¹ Reasons for the elevated risk remain unclear but may include a higher prevalence of risk factors, such as vitamin D deficiency because of low sun exposure; smoking; obesity; a diet high in smoked fish and marine mammals and low in fiber, fruits, and vegetables^101-103; and perhaps Helicobacter pylori infection.^104-106 The high CRC burden among Alaska Natives is compounded by inadequate availability of endoscopic services in much of Alaska,^{107, 108} resulting in the lowest CRC screening prevalence in the nation.¹⁰⁹

Racial disparities in CRC more broadly reflect differences in the prevalence of risk factors and access to high-quality, comprehensive health care. Much of this inequality stems from disproportionate wealth because of systemic racism.^{110, 111} In 2021, 24% of AIAN individuals and 20% of Black individuals lived below the poverty line compared with 9% of AAPI individuals and 8% of non-Hispanic White individuals.¹¹² Similarly, 31% of Native Hawaiian or other Pacific Islander individuals, 29% of AIAN individuals, and 19% of Black individuals aged 18–64 years were uninsured for at least part of 2021 compared with 12% of non-Hispanic White individuals. People with the lowest socioeconomic status are 40% more likely to be diagnosed with CRC than those with the highest socioeconomic status¹¹³ due in equal part to differences in the prevalence of CRC risk factors (e.g., smoking, obesity)¹¹⁴ and screening uptake.¹¹⁵

Stage at diagnosis plays the largest role in racial and ethnic survival disparities.¹¹⁶ Black individuals are most likely to be diagnosed with metastatic CRC (25% vs. 21% of non-Hispanic White individuals and 19% of AAPI individuals; Figure 7). Notably, the 5-year relative survival rate for localized disease is relatively similar (89%–91%) across racial and ethnic group (Figure 8). Later stage diagnosis reflects lower prevalence and quality of screening. For example, Black individuals are less likely to receive both timely follow-up of a positive stool test and high-quality colonoscopy.^{117, 118} Other contributors to mortality disparities include differences in the prevalence of comorbidities, unfavorable tumor characteristics, and inequality in receipt of high-quality treatment.^{82, 119, 120} One study estimated that one quarter of the Black–White survival disparity is explained by differences in tumor characteristics (e.g., grade, anatomic location) versus more than one half by differences in insurance status.¹²¹ A recent study of patients in the National Cancer Database found that, after accounting for stage, histology, and comorbidities, Black patients were 21% less likely to receive surgery for colon cancer and 28% less likely to receive surgery for rectal cancer.¹²² These differences accounted for 25% and 32% of the Black–White survival disparity for colon cancer and rectal cancer, respectively. Although higher mortality in Black individuals has been attributed in part to the disproportionate development of right-sided tumors,¹²¹ which are associated with lower survival,⁹¹ cancer registry data covering 99% of the population indicate that the distribution is similar (41%) to that of White individuals (40%), consistent with contemporary literature.^{123, 124}

The racial disparity in CRC mortality has slowly narrowed since 2005, when rates were almost 50% higher in Black individuals, to 33% in 2020 because the rapid declines of the 2000s have slowed in White individuals but persisted in Black individuals (Table 4). Among Hispanic and AAPI persons, mortality has declined steadily by about 2% per year since 2001 and 1990, respectively, but among AIAN persons, who have the highest rates, a downturn only began in 2011.

Geographic disparities

The striking variation in CRC globally reflects the large impact of lifestyle factors on cancer occurrence.¹²⁵ Similarly, wide differences within the United States in the prevalence of CRC risk factors, such as smoking and excess body weight, and access to high-quality health care, including screening, results in large geographic disparities. CRC incidence and mortality rates are lowest in the West and highest in Appalachia and parts of the South and Midwest. Rates range from 27.0 (per 100,000) in Utah to 46.5 in Mississippi for incidence and from 10.3 in Connecticut and 10.6 in Utah to 17.6 in Mississippi for mortality (Table 5). Geographic patterns are generally similar for Black and Whites individuals, particularly for mortality, highlighting the importance of socioeconomic status over race in cancer disparities.¹²⁶ CRC survival gaps are larger by race than by state.¹²⁷

TABLE 5. Colorectal cancer incidence (2015–2019) and mortality (2016–2020) rates and screening prevalence (2020) by state, United States.^a

	Incidence				Mortality				Screening (≥45 years)
	All races	NHW	NHB	Hispanic	All races	NHW	NHB	Hispanic	All races	NHW	NHB	Hispanic	45–54 years
Alabama	40.6	39.4	45.6	23.5	14.7	14.0	19.1	5.3	67	68	65	—	45
Alaska	39.5	33.3	29.8	28.0	14.9	12.5	—^b	—^b	63	61	—	71	39
Arizona	30.1	30.4	30.8	30.8	12.1	12.2	15.1	12.0	59	62	62	56	33
Arkansas	42.5	40.9	54.5	24.5	14.9	14.4	22.4	4.7	61	60	67	—	37
California	33.1	33.4	38.0	31.5	12.1	12.6	17.0	11.0	53	55	61	53	33
Colorado	30.5	29.4	36.1	34.9	11.3	10.9	17.0	13.4	63	65	57	55	36
Connecticut	32.6	31.9	37.7	34.8	10.3	10.4	12.3	8.3	67	68	63	64	41
Delaware	35.0	35.0	38.4	35.9	12.9	13.1	15.2	7.0	66	67	64	44	40
Dist. of Columbia	33.7	21.7	41.1	24.2	14.0	7.3	19.5	7.2	70	69	71	68	47
Florida	34.6	34.5	37.9	34.0	12.4	12.5	16.2	11.3	65	64	72	64	37
Georgia	38.6	37.3	43.7	35.7	14.1	13.5	17.2	7.8	64	62	69	50	38
Hawaii	37.8	36.4	40.9	55.3	11.7	12.7	12.8	16.0	65	65	—	64	38
Idaho	33.3	33.2	—^b	29.0	12.6	12.7	—^b	10.8	57	58	—	51	30
Illinois	39.4	39.0	48.4	32.9	14.0	13.7	20.8	9.5	59	61	59	52	36
Indiana	39.3	39.7	41.5	29.6	14.7	14.6	18.5	9.0	61	62	59	50	36
Iowa	40.1	40.2	51.1	31.2	13.6	13.8	20.6	7.1	63	64	62	36	36
Kansas	37.7	37.4	39.1	32.2	14.1	14.3	16.0	10.3	61	62	64	53	38
Kentucky	44.9	45.0	46.9	26.6	16.2	16.3	18.5	4.9	66	66	67	—	44
Louisiana	43.2	41.1	51.7	24.0	15.7	14.1	20.9	6.9	65	64	68	62	41
Maine	33.7	33.8	—^b	—^b	12.9	13.0	—^b	—^b	69	69	—	64	44
Maryland	34.3	34.3	38.0	21.7	13.3	12.8	16.5	6.0	67	65	72	59	42
Massachusetts	31.8	31.8	34.9	24.9	10.9	11.2	12.3	7.0	70	71	65	68	43
Michigan	35.1	34.1	43.6	27.4	13.4	13.0	18.3	9.3	66	66	66	72	38
Minnesota	34.8	34.4	37.7	32.9	11.8	11.7	12.8	8.7	65	65	67	52	38
Mississippi	46.5	43.1	54.7	23.7	17.6	16.2	21.4	5.6	62	61	64	—	39
Missouri	37.8	37.4	43.1	26.6	13.8	13.7	17.2	6.3	63	62	67	69	39
Montana	35.1	34.5	^b	28.3	12.0	11.7	—^b	11.0	59	59	—	62	33
Nebraska	39.8	39.8	42.3	35.1	14.3	14.6	17.5	8.7	62	63	62	43	38
Nevada^c	34.0	38.2	43.1	32.2	14.6	15.4	20.3	9.3	61	61	62	56	35
New Hampshire	33.6	33.8	^b	22.3	12.2	12.3	—^b	9.1	66	67	—	—	39
New Jersey	37.9	38.4	43.0	34.1	13.0	13.4	17.5	9.8	62	63	69	58	37
New Mexico	31.9	29.2	23.1	35.2	12.6	11.3	14.7	14.4	58	59	—	58	30
New York	35.9	35.9	38.5	32.9	11.7	12.0	14.6	9.2	67	68	68	67	44
North Carolina	34.5	33.6	38.3	27.4	12.6	12.1	16.2	7.6	66	66	70	49	40
North Dakota	39.0	37.7	—^b	—^b	13.1	12.8	—^b	—^b	63	63	—	—	41
Ohio	38.7	38.4	37.7	24.7	14.5	14.5	17.6	8.4	65	64	68	65	41
Oklahoma	39.6	37.5	42.2	32.9	16.3	16.0	21.5	9.0	56	58	61	34	31
Oregon	32.0	31.9	30.3	28.5	12.2	12.6	10.5	9.2	64	64	—	55	40
Pennsylvania	37.9	37.8	38.5	31.9	13.7	13.6	16.8	10.6	66	66	70	57	43
Rhode Island	31.1	30.8	24.8	23.7	11.6	12.2	8.2	5.8	69	71	68	66	45
South Carolina	35.5	34.2	40.6	23.1	13.4	12.5	17.8	7.5	67	66	71	—	42
South Dakota	38.5	38.2	—^b	—^b	14.3	13.9	—^b	—^b	65	66	—	47	39
Tennessee	38.3	37.8	43.9	21.1	14.8	14.4	19.5	6.1	65	66	68	—	38
Texas	36.6	36.6	45.0	35.5	13.7	13.8	19.6	12.3	59	64	61	53	32
Utah	27.0	26.3	32.4	33.6	10.6	10.6	12.8	9.9	64	66	—	58	37
Vermont	32.6	32.6	—^b	—^b	14.1	14.3	—^b	—^b	65	65	—	—	43
Virginia	32.9	32.6	37.7	21.3	13.2	12.9	17.6	6.4	66	66	70	63	41
Washington	32.7	32.2	38.8	28.5	11.9	12.2	14.9	8.1	63	64	57	59	36
West Virginia	43.7	44.0	42.0	—^b	16.6	16.7	17.4	—^b	63	63	58	—	42
Wisconsin	34.1	33.2	49.0	29.5	12.1	11.9	19.7	9.8	55	56	66	58	37
Wyoming	32.3	31.5	—^b	36.0	12.2	12.2	—^b	14.0	55	57	—	47	35
Puerto Rico^d									67	—	—	67	50
United States^e	35.9	35.7	41.7	32.5	13.1	13.1	17.6	10.7

• Note: Screening prevalence is age-adjusted to the 2000 US standard population; reflects a fecal occult blood test within the past year, or sigmoidoscopy within the past 5 years, or colonoscopy within the past 10 years; and does not distinguish between examinations for screening and diagnosis.
• Abbreviations: NHB, non-Hispanic Black; NHW, non-Hispanic White.
• ^a Rates are per 100,000 population and are age-adjusted to the 2000 US standard population. Incidence excludes appendiceal cancer.
• ^b Statistics are not displayed because there were fewer than 25 cases or 10 deaths.
• ^c Incidence data for Nevada are not included in US combined incidence rates because data did not meet inclusion standards for all years during 2015–2019 according to the North American Association of Central Cancer Registries. Incidence data for Nevada are based on data published in NAACCR's Cancer in North America, Volume II and include appendix.
• ^d Incidence data for Puerto Rico are not included in US combined rates for comparability to previously published data. Puerto Rico rates are not available by race/ethnicity. Mortality rates are for 2016 through 2020.
• ^e Screening prevalence for the United States is the median of state values.
• Source: Incidence: North American Association of Central Cancer Registries, 2022; Mortality: National Center for Health Statistics, 2022; Screening: Behavioral Risk Factor Surveillance System, 2020.

Colorectal cancer screening

As stated previously, CRC screening patterns have a large influence on CRC incidence and mortality trends. Observational studies suggest that colonoscopy reduces CRC incidence by about 40% and mortality by about 60%.^{41, 42, 128} The prevalence of reported up-to-date screening with any recommended test among individuals aged 50 years and older increased from 38% in 2000 to 66% in 2018 according to data from the NHIS.¹²⁹ In 2018, the American Cancer Society lowered the recommended age to begin screening in average-risk individuals from 50 to 45 years based on empirical evidence of increasing risk at younger ages and microsimulation modeling that showed greater benefit than harm.^{19, 130} The US Preventive Services Task Force reached the same conclusion in their recommendation update in 2021,^{131, 132} and studies published since have supported the earlier screening. Screening average-risk individuals aged 45–49 years has only had a modest impact on colonoscopy volume and results in the same prevalence of advanced adenoma (6.3%) and sessile serrated lesions (2.9%) as in those aged 50–54 years (5.8% and 3.0%, respectively).¹³³ In data from the Nurses' Health Study II, screening initiation at ages 45–49 years compared with 50–54 years resulted in an absolute reduction in the cumulative incidence of CRC of 72 per 100,000 women by age 60 years.¹³⁴ A retrospective cohort study in Florida found that screening at ages 45–49 years reduced the risk of CRC incidence by 50% compared with no screening at a median follow-up of 7 years.¹³⁵

According to the NHIS, 59% of individuals aged 45 years and older were up to date on CRC screening in 2021, ranging from 50% of Asian individuals to 61% of White and Black individuals (Table 6). However, uptake remains low in individuals aged 45–49 years (20%) and 50–54 years (50%). Other characteristics associated with low screening prevalence include residence in the United States for <10 years (29%), being uninsured (21%), and having less than a high school education (48%). Screening also varies widely by state, with prevalence in 2020 ranging from 53% in California to 70% in the District of Columbia and Massachusetts (Figure 9 and Table 5).

TABLE 6. Colorectal cancer screening (%), adults aged 45 years and older, 2021, United States.

	Stool test^a	Colonoscopy^b	Up to date^c
	≥45 years	≥45 years	≥45 years	45–75
Overall	10	54	59	58
Sex
Males	9	54	58	56
Females	10	55	60	60
Age (years)
45–49	3	18	20	—
50–54	9	43	50	—
55–64	11	65	70
65–75	—	—	—	83
65–74	15	74	80	—
≥75	10	67	70	—
Race/ethnicity
Hispanic	14	46	52	51
White only	9	57	61	60
Black only	11	57	61	59
Asian only	10	45	50	48
AIAN only or multiple	10	48	52	52
Sexual orientation
Gay/lesbian	12	57	64	61
Straight	10	55	59	58
Bisexual	—	48	51	57
Immigration status
Born in United States/US Territory	9	57	61	60
In United States <10 years	9	25	29	30
In United States ≥10 years	12	48	53	52
Education
Less than high school	11	43	48	47
High school diploma	9	51	55	54
Some college	11	56	61	59
College graduate	9	60	64	63
Income level
<100% FPL	11	42	47	46
100% to <200% FPL	12	47	52	51
≥200% FPL	9	58	62	61
Insurance status
Uninsured	4	18	21	22
Private	9	59	63	64
Medicaid/Public/dual eligible	11	48	52	53
Medicare (ages ≥65 years)	15	69	75	82
Other	15	68	73	74

• Note: Estimates do not distinguish between examinations for screening and diagnosis. All estimates except for age and insurance status are age-adjusted to the 2000 US standard population.
• Abbreviations: AIAN, American Indian/Alaska Native; FPL, federal poverty level.
• ^a Fecal occult blood test (FOBT), or fecal immunochemical test (FIT) within the past 1 year, or stool DNA (sDNA) test within the past 3 years.
• ^b Within the past 10 years.
• ^c For ages ≥45 and ≥50 years: FOBT/FIT, sDNA, sigmoidoscopy, colonoscopy, computed tomography (CT) or colonography, in the past 1, 3, 5, or 10 years, respectively. For ages 50–75 years: FOBT/FIT, sigmoidoscopy, colonoscopy, CT colonography, or sDNA test in the past 1, 5, and 10 years and the past 5 and 3 years, respectively, or sigmoidoscopy in the past 10 years with FOBT/FIT in the past 1 year.
• Source: National Health Interview Survey, 2021.

FIGURE 9

Up-to-date colorectal cancer screening (%), adults aged 45 years and older by state, 2020, United States. Colorectal screening consisted of fecal blood test, sigmoidoscopy, or colonoscopy within the past 1, 5, and 10 years, respectively. Prevalence is age-adjusted to the 2000 US standard population and does not distinguish between screening and diagnostic examinations. Source: Behavioral Risk Factors Surveillance System, 2020.

The coronavirus disease 2019 pandemic caused major disruptions in access to preventive health care like cancer screening. Compared with 2019, CRC screening rates during the second quarter of 2020 plummeted by approximately 80%.¹³⁶ However, a recent study based on BRFSS data found that, in contrast to declines from 2018 to 2020 for breast and cervical cancer screening, CRC screening remained steady overall because a 16% decline in colonoscopy was counterbalanced by a 7% increase in stool testing.¹³⁷ The increase in stool testing highlights the opportunity for these noninvasive options to both raise screening prevalence, especially among underserved communities who prefer home-based options, and help avoid downturns in screening during future health care disruptions.^{138, 139} It has been estimated that achieving 80% CRC screening prevalence nationally would prevent tens of thousands of CRC cases and deaths.¹⁴⁰ The multitarget stool DNA test is an advance in home-based alternatives because, with real-world adherence, it results in more life-years gained than fecal immunochemical testing (FIT) or high-sensitivity guaiac-based fecal occult blood testing.¹⁴¹ Multitarget stool DNA testing is also more sensitive than fecal immunochemical testing or colonoscopy alone for serrated polyps,¹⁴² which are precursors for 15%–30% of all CRCs, including many interval cancers.^{143, 144} However, although the 3-year screening schedule for multitarget stool DNA testing may be less burdensome than annual testing, the cost remains substantially higher compared to other stool tests.¹⁴⁵

CONCLUSIONS

Although overall CRC mortality continues to decline, this progress is tempered by a rapidly changing landscape of disease that foreshadows less favorable trends ahead. First, the CRC burden is shifting to younger individuals as cohorts born in the last one half of the 20th century who have elevated risk age; one in five new cases now occur in individuals in their early 50s or younger. Second, there is an overall shift to later stage disease, with more individuals now diagnosed at an advanced stage than in the mid-1990s before widespread screening. Finally, there is a shift from right-sided to left-sided tumors, despite higher efficacy for preventing the latter through screening, likely reflecting changes in underlying disease risk of unknown etiology. In addition, striking disparities by race and geography persist, with mortality rates in Alaska Natives almost three times higher than those in non-Hispanic White individuals [Correction added on 30 March 2023, after first online publication: In preceding sentence "...Alaska Natives almost four times higher..." has been changed to "...Alaska Natives almost three times higher...." in this version.]. Although a substantial proportion of CRC deaths can be prevented through screening, four in 10 Americans aged 45 years and older are not up to date nationally, including one half of Californians. Screening is especially low among younger individuals and those without health insurance. Reducing CRC inequalities and furthering progress could be achieved by incentivizing healthier lifestyles and ensuring equitable access to high-quality health care for all individuals, especially those in rural and other low-resource areas, such as Alaska. In addition, research is needed to elucidate causes for rising CRC incidence and to advance treatment options for tumor subtypes without effective therapies.

ACKNOWLEDGMENTS

The authors gratefully acknowledge all cancer registries and their staff for their hard work and diligence in collecting cancer information, without which this research could not have been done.

CONFLICT OF INTEREST STATEMENT

Andrea Cercek reports grants and contracts from GlaxoSmithKline and Seagen Inc.; and personal fees from Bayer, G1 Therapeutics, GlaxoSmithKline, Janssen Biotech, Merck, Pfizer Pharmaceuticals LLC, and Seagen Inc. outside the submitted work. The other authors declare no conflicts of interest.