Skip to content

Epidemiology and burden

TL;DR — Colorectal cancer is the third most frequently diagnosed cancer and second leading cause of cancer death globally; GLOBOCAN 2022 attributed 9.6% of new cancers and 9.3% of cancer deaths to the colorectum (Bray 2024, PMID 38572751). GLOBOCAN estimated 1.9 million cases and 930,000 deaths in 2020 and projected 3.2 million cases and 1.6 million deaths in 2040 under demographic change (Morgan 2023, PMID 36604116). Incidence is highest in many high-HDI populations, but mortality increasingly concentrates where screening and treatment access lag. Declining rates in screened older adults coexist with rising birth-cohort risk below 50 and now into ages 50–54 (Gupta 2024, PMID 38081492). “Disparity” is not a single covariate: stage, insurance, rurality, education, treatment and structural access interact, so descriptive racial or geographic differences should not be read as biological causation (Islami 2024, PMID 37962495).

What the major estimates measure

Source framework Numerator Denominator/model Principal strength Principal limitation
GLOBOCAN Incident cases and deaths in a reference year Registries plus modelling across 185 countries Globally comparable snapshot Data quality and modelling intensity vary (Bray 2024, PMID 38572751)
GBD Incidence, deaths, YLLs, YLDs and DALYs Cause-of-death and disease modelling across 204 locations Burden and uncertainty intervals Estimates differ from GLOBOCAN by methods and case definitions (Kocarnik 2022, PMID 34967848)
National registries Recorded incident cancers, stage and survival Defined resident population Time trends and subgroup detail Registration completeness and stage coding vary
Claims/hospital datasets Diagnoses, procedures and outcomes Covered or treated population Care pathway and utilization Not population incidence; coverage selection
Screening cohorts Screen-detected lesions and interval cancers Invited or participating population Program performance Participation and healthy-user selection

Counts, age-standardized rates and lifetime risk answer different questions. Counts rise with population growth and aging even when age-standardized rates fall. Cross-country comparisons should use age-standardized rates; health-system planning needs absolute counts.

Global burden

GLOBOCAN 2022 estimated almost 20 million cancers and 9.7 million cancer deaths overall. Colorectal cancer represented 9.6% of incident cancers and 9.3% of deaths, ranking behind lung and breast for incidence and behind lung for mortality (Bray 2024, PMID 38572751).

For colorectal cancer specifically, GLOBOCAN 2020 estimated more than 1.9 million cases and 930,000 deaths. Under demographic projections, 2040 burden reached 3.2 million cases and 1.6 million deaths (Morgan 2023, PMID 36604116).

GBD 2019 estimated 2.17 million incident colorectal cancers (95% UI 2.00–2.34 million), 1.09 million deaths (95% UI 1.00–1.15 million) and 24.3 million DALYs (95% UI 22.6–25.7 million). Its estimates differ from GLOBOCAN by year, inputs and modelling framework and should be displayed beside, not averaged with, GLOBOCAN estimates (Global Burden of Disease 2019 Cancer Collaboration 2022, PMID 34967848).

Metric Estimate Year Framework Interpretation
Incident colorectal cancers 1.9 million 2020 GLOBOCAN Registry/model estimate (Morgan 2023, PMID 36604116)
Colorectal-cancer deaths 930,000 2020 GLOBOCAN Same framework as above
Projected incident cancers 3.2 million 2040 GLOBOCAN demographic projection Not a forecast of future policy or screening
Projected deaths 1.6 million 2040 GLOBOCAN demographic projection Sensitive to future treatment and prevention
Incident colorectal cancers 2.17 million (95% UI 2.00–2.34) 2019 GBD Different input/model; not directly interchangeable (PMID 34967848)
Deaths 1.09 million (95% UI 1.00–1.15) 2019 GBD Different input/model (PMID 34967848)
DALYs 24.3 million (95% UI 22.6–25.7) 2019 GBD Years lost plus disability (PMID 34967848)

Geographic variation

In 2020, age-standardized incidence was highest among men in Australia/New Zealand and European regions (40.6 per 100,000) and lowest among women in several African regions and Southern Asia (4.4 per 100,000). Mortality ranged from 20.2 per 100,000 among men in Eastern Europe to 2.5 among women in Southern Asia (Morgan 2023, PMID 36604116).

High incidence can coexist with lower case-fatality where screening detects earlier disease and treatment is accessible. Conversely, low recorded incidence can coexist with high mortality-to-incidence ratios where diagnosis is late or registration incomplete. Socioeconomic inequalities are visible at the international level even after age standardization (Cao 2024, PMID 38616547).

Country profiles can move in opposite directions within the same period. Comparative China–United States statistics show how rapid incidence change, aging, registry coverage and different screening maturity create dissimilar burden trajectories despite convergence in some exposures (Xia 2022, PMID 35143424). Regional labels therefore conceal substantial within-country variation.

Westernization is a shorthand, not a mechanism. It bundles dietary energy density, processed/red meat, obesity, inactivity, alcohol, smoking, microbiome-altering exposures, longevity and diagnostic capacity (Keum 2019, PMID 31455888). Analyses should avoid assigning observed national change to any single exposure without individual-level longitudinal data.

Age and sex

Absolute incidence rises steeply with age, which is why population aging drives case counts. Men generally have higher age-standardized incidence and mortality than women, but the size of the difference varies by site and region (Morgan 2023, PMID 36604116).

Tumor location also changes with age and sex. Proximal cancers are more common in older adults and women; rectal cancers feature prominently in early-onset trends. These distributions matter because stool-test sensitivity, endoscopic protection and molecular subtype differ by site.

Older adults are underrepresented in many randomized trials. Chronologic age does not directly measure frailty, competing mortality or treatment tolerance; population survival comparisons by age are therefore mixtures of biology, stage, treatment selection and competing risk.

Early-onset and birth-cohort colorectal cancer

Early-onset colorectal cancer is conventionally diagnosed before age 50. The threshold is administrative and historical rather than a biological boundary.

Incidence under 50 has increased in multiple high-income countries since the 1990s. One synthesis reported roughly 2% annual increase since 1994 and enrichment for left-sided/rectal, advanced-stage, poorly differentiated and signet-ring features, while acknowledging selection and changing diagnosis (Mauri 2019, PMID 30520562).

A more useful frame is birth-cohort colorectal cancer: successively higher risk among people born from approximately 1960 onward, increasing disease at ages 50–54, flattening prior declines at 55–74, and larger increases for rectal and distant-stage disease (Gupta 2024, PMID 38081492). This frame prevents the false inference that the phenomenon stops at the 50th birthday.

Observation Quantitative/empirical anchor What remains unknown
Rising incidence below 50 Reported across countries and both sexes (Spaander 2023, PMID 37105987) Relative contribution of exposure, detection and classification
Projected 2030 share 11% of colon and 23% of rectal cancers Projection uncertainty and screening response
Hereditary fraction About 20% in the 2023 primer Which germline testing strategy is optimal
Birth-cohort extension Increasing 50–54 and flattening 55–74 trends Whether exposure windows begin in utero, childhood or adulthood (Gupta 2024, PMID 38081492)
Later stage at diagnosis Recurrent observation in young-onset cohorts Symptom delay versus aggressive biology versus absent screening

Candidate explanations include obesity/metabolic disease, diet, antibiotics, microbiome change, sedentary behavior and environmental exposures. None currently explains the full international, anatomic and cohort pattern (Saraiva 2023, PMID 36925459; Spaander 2023, PMID 37105987).

Young-onset disease also carries a different burden profile: fertility, sexual function, employment, dependent care and decades of survivorship. A stage-equivalent survival comparison does not capture years of life lost or financial toxicity.

Modifiable and medical risk factors

Exposure/domain Direction of association Causal confidence and caveat
Adiposity Higher risk Supported across observational and mechanistic evidence; residual confounding remains (Keum 2019, PMID 31455888)
Physical inactivity Higher risk Correlated with adiposity and metabolic health
Alcohol Dose-related higher risk Intake measurement is imprecise
Smoking Higher incidence and mortality Stronger for some molecular/anatomic subtypes
Processed/red meat patterns Higher risk Food substitution and total dietary pattern matter
Fiber/whole grains Lower risk association Measurement and healthy-user confounding
Inflammatory bowel disease Higher duration/extent-dependent risk Modern inflammation control may alter historic estimates
Type 2 diabetes Higher risk association Metabolic and treatment confounding
Aspirin Lower incidence in selected settings Bleeding risk prevents universal inference; strongest randomized prevention evidence in Lynch syndrome (Burn 2020, PMID 32534647)

Population-attributable fractions depend on exposure prevalence, relative risks and causal assumptions. They are scenario models, not fractions of individual cancers that can be assigned a cause.

Hereditary and familial burden

Lynch syndrome and APC-associated polyposis are the best-characterized high-penetrance syndromes, but broader panels identify pathogenic variants across multiple DNA-repair and polyposis genes. Universal-panel cohorts found pathogenic variants in 14.2% of 34,244 commercially tested patients and 15.5% of 361 prospectively unselected patients; cohort selection, panel content and variant classification affect yield (Coughlin 2022, PMID 36370464; Uson 2022, PMID 33857637).

Familial aggregation without an identified high-penetrance variant still matters. Family history can encode shared polygenic risk, shared exposure and undetected variants. Epidemiologic categories should not collapse “no pathogenic variant found” into “nonfamilial.”

Stage at diagnosis

Stage distribution is a health-system outcome as well as tumor biology. Screening increases detection of precursors and earlier cancers; access delays, symptom interpretation and referral pathways shift presentation later.

SEER-linked analysis associated race, insurance, poverty, language isolation and unemployment with late-stage presentation, illustrating multiple linked mechanisms rather than a single “patient factor” (Patel 2019, PMID 30788664). Adults with disabilities face additional barriers across detection, diagnosis and treatment (Iezzoni 2022, PMID 35358465).

Emergency presentation is a high-risk pathway. Obstruction or perforation is associated with more advanced disease, physiological compromise and higher perioperative risk; comparisons with elective presentation are confounded by stage and selection (Ogawa 2022, PMID 35847445).

Racial, socioeconomic and rural disparities

In the United States, overall cancer mortality was approximately 1.6–2.8 times higher among people with ≤12 versus ≥16 years of education within Black and White groups. For colorectal cancer specifically, mortality in nonmetropolitan areas was 23% higher among men and 21% higher among women than in large metropolitan areas (Islami 2024, PMID 37962495).

Axis Measured pathway Avoidable misinterpretation
Race/ethnicity Exposure to structural inequity, screening, stage, care and comorbidity Race as a genetic proxy
Education/income Insurance, transport, time, health literacy, food environment Individual blame
Rurality Travel, specialist density, pathology/surgery volume, trial access Rural residence as homogeneous
Disability Inaccessible screening equipment and diagnostic/treatment barriers Assuming standard pathways are equally usable (Iezzoni 2022, PMID 35358465)
Language/immigration Communication, trust and navigation Treating language as preference rather than access infrastructure

Race and rurality can interact. In a National Cancer Database analysis of 463,948 stage II–III cases, the joint exposure of Black race and rural residence was associated with the worst survival after adjustment, although residual confounding and database selection remain (Tobin 2023, PMID 36890731). Geographic survival disparities track persistent poverty as well as metropolitan status (Coughlin 2025, PMID 40422507).

Longitudinal Medicare data show that geographic and sociodemographic disparities changed over 1973–2010 rather than remaining fixed, supporting policy and health-system causation over immutable group differences (Liang 2017, PMID 27578387). Rural residence is also associated with later stage for screen-preventable cancers in registry analyses (Zahnd 2018, PMID 29550163).

Screening behavior is shaped by beliefs and supply simultaneously. National survey analysis found geographic variation in screening and cancer fatalism, but cross-sectional association cannot establish whether beliefs cause nonparticipation or reflect lived access barriers (Moss 2019, PMID 31615890).

Screening disparities are intervention points, not fixed traits. Organized invitation, mailed FIT, navigation, reminders and low-barrier diagnostic colonoscopy act at different steps; overviews emphasize multilevel rather than education-only interventions (Huang 2017, PMID 29546218).

Survival and mortality

Population survival is driven by stage, tumor biology, treatment, competing mortality and access. Improvements over time may reflect screening-induced stage shift, better surgery, adjuvant therapy and metastatic treatments, but lead-time and overdiagnosis complicate interpretation.

Cancer-specific mortality avoids some competing-risk effects but depends on cause-of-death attribution. Overall survival is unambiguous but can obscure cancer-control differences in older or comorbid populations.

DALYs combine premature mortality and disability. In cancer, years of life lost dominate; early-onset disease therefore contributes disproportionately per death even when it remains a minority of cases (Kocarnik 2022, PMID 34967848).

COVID-19 disruption as a natural experiment

Population screening paused or contracted during early pandemic waves. Catalonia documented short-term disruption in invitation, testing and diagnostic pathways, followed by recovery efforts (Vives 2022, PMID 34954239). Screening-program analyses observed altered diagnosis volume and stage pathways, but attribution requires separating delayed diagnosis from secular change (Cubiella 2023, PMID 37835547).

The episode illustrates that screening benefit is a chain: invitation → completion → positive-test colonoscopy → treatment. A program can report high test sensitivity while losing benefit through delays downstream.

Measurement cautions

  1. Colon and rectal cancers may be pooled despite different trend slopes.
  2. Age-standardized rates depend on the chosen standard population.
  3. Registry incidence reflects diagnostic capacity as well as disease.
  4. Early-onset proportions can rise because older-age incidence falls.
  5. Race/ethnicity categories differ across jurisdictions and time.
  6. Survival comparisons are distorted by lead time and stage migration.
  7. GLOBOCAN and GBD estimates should never be averaged.

Open questions

  • Which early-life exposures reproduce the birth-cohort, rectal-predominant and international pattern of early-onset disease? (Gupta 2024, PMID 38081492)
  • What fraction of projected 2040 burden is preventable by feasible risk-factor and screening interventions in low- and middle-income settings? (Morgan 2023, PMID 36604116)
  • Which multilevel interventions close rural mortality gaps rather than only raise one-time screening completion? (Tobin 2023, PMID 36890731)
  • How should disability-access measures be incorporated into screening quality reporting? (Iezzoni 2022, PMID 35358465)
  • Can surveillance systems distinguish true incidence change from diagnostic and coding change quickly enough to guide policy? (Bray 2024, PMID 38572751)

References

  1. Bray F, et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2024;74(3):229-263. PMID 38572751
  2. Morgan E, et al. Global burden of colorectal cancer in 2020 and 2040: incidence and mortality estimates from GLOBOCAN. Gut. 2023;72(2):338-344. PMID 36604116
  3. Gupta S, et al. Birth Cohort Colorectal Cancer (CRC): Implications for Research and Practice. Clin Gastroenterol Hepatol. 2024;22(3):455-469.e7. PMID 38081492
  4. Islami F, et al. American Cancer Society's report on the status of cancer disparities in the United States, 2023. CA Cancer J Clin. 2024;74(2):136-166. PMID 37962495
  5. Global Burden of Disease 2019 Cancer Collaboration, et al. Cancer Incidence, Mortality, Years of Life Lost, Years Lived With Disability, and Disability-Adjusted Life Years for 29 Cancer Groups From 2010 to 2019: A Systematic Analysis for the Global Burden of Disease Study 2019. JAMA Oncol. 2022;8(3):420-444. PMID 34967848
  6. Cao W, et al. Socioeconomic inequalities in cancer incidence and mortality: An analysis of GLOBOCAN 2022. Chin Med J (Engl). 2024;137(12):1407-1413. PMID 38616547
  7. Xia C, et al. Cancer statistics in China and United States, 2022: profiles, trends, and determinants. Chin Med J (Engl). 2022;135(5):584-590. PMID 35143424
  8. Keum N, Giovannucci E. Global burden of colorectal cancer: emerging trends, risk factors and prevention strategies. Nat Rev Gastroenterol Hepatol. 2019;16(12):713-732. PMID 31455888
  9. Mauri G, et al. Early-onset colorectal cancer in young individuals. Mol Oncol. 2019;13(2):109-131. PMID 30520562
  10. Spaander MCW, et al. Young-onset colorectal cancer. Nat Rev Dis Primers. 2023;9(1):21. PMID 37105987
  11. Saraiva MR, Rosa I, Claro I. Early-onset colorectal cancer: A review of current knowledge. World J Gastroenterol. 2023;29(8):1289-1303. PMID 36925459
  12. Burn J, et al. Cancer prevention with aspirin in hereditary colorectal cancer (Lynch syndrome), 10-year follow-up and registry-based 20-year data in the CAPP2 study: a double-blind, randomised, placebo-controlled trial. Lancet. 2020;395(10240):1855-1863. PMID 32534647
  13. Coughlin SE, et al. Multigene Panel Testing Yields High Rates of Clinically Actionable Variants Among Patients With Colorectal Cancer. JCO Precis Oncol. 2022;6:e2200517. PMID 36370464
  14. Uson PLS, et al. Germline Cancer Susceptibility Gene Testing in Unselected Patients With Colorectal Adenocarcinoma: A Multicenter Prospective Study. Clin Gastroenterol Hepatol. 2022;20(3):e508-e528. PMID 33857637
  15. Patel A, et al. The role of socioeconomic disparity in colorectal cancer stage at presentation. Updates Surg. 2019;71(3):523-531. PMID 30788664
  16. Iezzoni LI. Cancer detection, diagnosis, and treatment for adults with disabilities. Lancet Oncol. 2022;23(4):e164-e173. PMID 35358465
  17. Ogawa K, et al. Evaluation of clinical outcomes with propensity-score matching for colorectal cancer presenting as an oncologic emergency. Ann Gastroenterol Surg. 2022;6(4):523-530. PMID 35847445
  18. Tobin EC, et al. The Intersection of Race and Rurality and its Effect on Colorectal Cancer Survival. Am Surg. 2023;89(7):3163-3170. PMID 36890731
  19. Coughlin SS, et al. The Influence of Poverty and Rurality on Colorectal Cancer Survival by Race/Ethnicity: An Analysis of SEER Data with a Census Tract-Level Measure of Persistent Poverty. Curr Oncol. 2025;32(5). PMID 40422507
  20. Liang PS, et al. Temporal Trends in Geographic and Sociodemographic Disparities in Colorectal Cancer Among Medicare Patients, 1973-2010. J Rural Health. 2017;33(4):361-370. PMID 27578387
  21. Zahnd WE, Fogleman AJ, Jenkins WD. Rural-Urban Disparities in Stage of Diagnosis Among Cancers With Preventive Opportunities. Am J Prev Med. 2018;54(5):688-698. PMID 29550163
  22. Moss JL, et al. Geographic disparities in cancer screening and fatalism among a nationally representative sample of US adults. J Epidemiol Community Health. 2019;73(12):1128-1135. PMID 31615890
  23. Huang JL, et al. Approaching the Hard-to-Reach in Organized Colorectal Cancer Screening: an Overview of Individual, Provider and System Level Coping Strategies. AIMS Public Health. 2017;4(3):289-300. PMID 29546218
  24. Vives N, et al. Short-term impact of the COVID-19 pandemic on a population-based screening program for colorectal cancer in Catalonia (Spain). Prev Med. 2022;155:106929. PMID 34954239
  25. Cubiella J, et al. Impact of the COVID-19 Pandemic on the Diagnosis of Colorectal Cancer within a Population-Based Organized Screening Program. Cancers (Basel). 2023;15(19). PMID 37835547