Comparative analysis of outcomes and sociodemographic factors in early- and late-onset colorectal cancer hospitalizations

Introduction

Colorectal cancer (CRC) is the third most commonly diagnosed cancer worldwide and remains a major contributor to cancer-related morbidity and mortality (1). While the overall incidence of CRC in older adults has declined due to widespread screening efforts, the incidence of early-onset CRC (EOCRC), defined as CRC diagnosed before age 50, has steadily increased by approximately 2.4% between 2012 and 2021 (2). In 2019, 20% (1 in 5) of CRCs were in people 54 years or younger, up from 11% (1 in 10) in 1995 (1). The rising incidence of EOCRC has prompted changes in screening guidelines. The American Cancer Society now recommends that individuals begin screening for CRC at age 45, down from the previous recommendation of age 50, in response to the increasing number of younger patients being diagnosed (3).

Early-onset CRC is distinct in its clinical behavior, often presenting at more advanced stages and with more aggressive histologic features compared to late-onset CRC (LOCRC), defined as CRC at or above the age of 50 (4). Recent large-scale studies have further demonstrated that EOCRC harbors a distinct molecular fingerprint with unique genomic alterations and worse prognosis compared to LOCRC (5). Genetic predispositions such as Lynch syndrome and familial adenomatous polyposis contribute to some cases, but growing evidence implicates modifiable factors such as obesity in driving this trend (6,7).

Obesity is a well-established risk for CRC. Both overall body size and weight gain during adulthood have been associated with an increased risk of CRC, further supporting the role of modifiable lifestyle factors in CRC development (6,8,9). In particular, long-term weight gain in adulthood has been shown to independently increase CRC risk, highlighting the importance of weight management throughout life (8). A large meta-analysis showed that each 5 kg/m² increase in body mass index (BMI) was associated with an 18% increased risk of colon cancer in men and 11% in women, with a stronger link to colon than rectal cancer (10). Obesity-related CRC is more often located in the proximal colon and arises via complex mechanisms, including chronic low-grade inflammation, insulin resistance, and alterations in gut microbiota (11). Furthermore, central (visceral) obesity may be even more predictive of CRC risk than BMI alone (11,12). In a recent 2018 prospective cohort study, women with obesity were found to have a 93% higher risk of EOCRC compared to those with normal BMI, even after controlling for dietary and lifestyle factors (13). These findings suggest that excess adiposity during early adulthood is an independent risk factor for EOCRC and may partially explain the shifting age distribution of the disease.

Racial, ethnic, and gender disparities compound this public health challenge. African Americans are disproportionately affected by EOCRC and often experience delayed diagnosis and worse outcomes. Similarly, EOCRC appeared to affect a greater number of Hispanic patients younger than 40 years (14,15). These inequities are shaped by structural barriers to care, environmental exposures, and socioeconomic determinants of health, and they underscore the need for population-specific prevention strategies.

From a clinical standpoint, EOCRC is associated with more intensive treatment regimens, fertility preservation concerns, and increased hospitalizations (16,17). A recent analysis of the National Inpatient Sample (NIS) (2016–2021) examined trends and inpatient mortality risk factors in early-onset colon cancer hospitalizations among nonelective admissions, but did not include rectal cancer, elective admissions, or a direct comparison with LOCRC (18).

The present study aims to directly compare sociodemographic characteristics, comorbidity profiles, inpatient chemotherapy utilization, length of stay (LOS), and inpatient mortality between EOCRC and LOCRC hospitalizations using the NIS from 2016 to 2020. We present this article in accordance with the STROBE reporting checklist (available at https://jgo.amegroups.com/article/view/10.21037/jgo-2026-0285/rc).

Methods

We performed a retrospective analysis using population-level data from the NIS database among patients hospitalized with CRC between 2016 and 2020. The NIS is a hospitalization-level database, and individual patients may contribute multiple admissions across the study period (18). The study cohort included hospitalized patients with an ICD-10-CM diagnosis code for CRC during the hospitalization. CRC hospitalizations were identified using primary ICD-10-CM diagnosis codes for colon cancer (C18), rectosigmoid junction (C19), and rectal cancer (C20). Metastatic disease codes were not included in the identification algorithm. Patients were excluded if they were younger than 18 years. Out of 34 million hospitalizations from 2016–2020, a total of 246,231 patients met inclusion criteria. We further classified them as EOCRC (age <50) or LOCRC (age ≥50) based on age at the time of hospitalization, not age at diagnosis.

The primary outcome of the study was to compare sociodemographics (age, race, sex, insurance type) among patients hospitalized with EOCRC vs. LOCRC. Secondary outcomes were to assess use of inpatient chemotherapy, role of BMI and obesity, LOS, and inpatient mortality.

Multivariable logistic regression was performed to evaluate associations with EOCRC vs. LOCRC hospitalizations. Sociodemographic characteristics including sex, race/ethnicity, hospital setting, insurance type, BMI categories, and inpatient mortality were examined in the first analysis. A separate analysis examined comorbidities and clinical characteristics including obesity (ICD-coded), depression, tobacco use disorder, alcohol use disorder, diabetes mellitus, coronary artery disease (CAD), cerebrovascular accident, chronic kidney disease, end stage renal disease, hypertension, inpatient cardiac arrest, and inpatient chemotherapy utilization (using ICD-10-PCS procedure codes). BMI categories were analyzed separately from ICD-coded obesity to avoid collinearity.

The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This study used publicly available de-identified data and was exempt from Institutional Review Board (IRB) review. Statistical analyses were performed using Stata, version 18.0 (StataCorp, College Station, TX) (20).

Results

A total of 246,231 inpatient hospitalizations with a primary diagnosis of CRC were identified from the 2016–2020 NIS. Of these, 33,662 (13.7%) were classified as EOCRC, while 212,569 (86.3%) were classified as LOCRC. The mean age at admission was 42.9 years [95% confidence interval (CI): 42.88–43.01 years] for EOCRC and 69.5 years (95% CI: 69.49–69.59 years) for LOCRC. Table 1 summarizes the unadjusted baseline characteristics of EOCRC and LOCRC hospitalizations across the full study cohort.

Multivariate logistic regression was used to evaluate the relationship between sociodemographic factors, hospitalization characteristics, and BMI in patients with EOCRC and LOCRC (Table 2). More female patients with EOCRC were hospitalized compared to male patients [odds ratio (OR): 1.09, P<0.001]. There was no significant difference in mean hospital LOS between the two groups, with early-onset CRC patients averaging 6.01 days and LOCRC patients averaging 6.66 days. Patients with EOCRC were noted to have a decreased likelihood of inpatient mortality (OR: 0.649, P<0.001) whereas those with LOCRC were more likely to have an increased likelihood (OR: 1.539, P<0.001). Patients diagnosed with EOCRC were more likely to have Medicaid, private commercial insurance or no insurance. Those with LOCRC were most likely to have Medicare be the primary payor.

When evaluating racial and ethnic distribution in hospitalized patients with EOCRC, Black (OR: 1.17, P<0.001), Hispanic (OR: 1.56, P<0.001), and Asian (OR: 1.29, P<0.001) patients were more likely to have EOCRC when compared to White patients. In contrast, for LOCRC, all non-White racial groups had a slightly decreased likelihood of CRC compared to White patients (P<0.001 for all groups).

BMI was used as an objective measure of body size and was included when analyzing overall patient characteristics. However, due to a high proportion of missing BMI values (75.7%), BMI was not incorporated into multivariable regression models evaluating comorbidities. Instead, a diagnosis of obesity, captured through ICD codes, was used as a proxy for clinician-recognized obesity in regression analyses. This allowed for more complete patient inclusion and reflected obesity as a coded comorbidity. Among patients with available BMI data, concordance with obesity diagnosis was high in the upper BMI strata: 85.2% of patients with BMI 30–39 and 88.7% with BMI ≥40 had an associated obesity diagnosis. In contrast, only 2.2% of patients with missing BMI values were diagnosed with obesity, underscoring differences in recognition and coding behavior. Given these discrepancies, BMI and obesity were analyzed separately to preserve their distinct interpretative roles and avoid collinearity.

Both BMI and obesity were significantly associated with EOCRC hospitalizations. A diagnosis of obesity was independently associated with increased odds of EOCRC (OR: 1.54, P<0.001). Stratified by BMI, patients with BMI 30–39 had a modestly elevated risk (OR: 1.07, P=0.02), and those with BMI ≥40 had substantially increased odds (OR: 1.50, P<0.001), compared to those with BMI <20. Interestingly, patients with BMI 20–29 had a slightly decreased likelihood of EOCRC admissions (OR: 0.93, P=0.02). Among patients with LOCRC, a different pattern emerged: BMI 20–29 was associated with a slightly increased likelihood of admissions (OR: 1.07, P=0.02), while BMI 30–39 (OR: 0.94, P=0.02) and BMI ≥40 (OR: 0.67, P<0.001) were associated with decreased odds of LOCRC.

Further analysis was completed to understand the role of comorbidities associated with EOCRC and LOCRC as well as inpatient interventions such as chemotherapy (Table 3). Chemotherapy use during hospitalization was more common among EOCRC patients compared to LOCRC (OR: 1.78, P<0.001 vs. OR: 0.56, P<0.001, respectively). Comorbidities that had significant association with EOCRC included concomitant diagnostic codes for obesity as discussed above, depression (OR: 1.13, P<0.001) and tobacco use disorder (OR: 1.26, P<0.001). Patients admitted with EOCRC were less likely to have chronic medical conditions such as diabetes, CAD, anemia, or kidney disease. Interestingly, patients with EOCRC were also less likely to have alcohol use disorder (OR: 0.86, P<0.001). Patients admitted with LOCRC had an increased odds of having chronic medical conditions such as diabetes, CAD, and kidney disease and associated with an increased likelihood of alcohol use disorder (OR: 1.17, P<0.001). Interestingly, they were less associated with a diagnosis of obesity (OR: 0.65, P<0.001), tobacco use disorder (OR: 0.79, P<0.001) and depression (OR: 0.88, P<0.001).

Discussion

This study successfully characterized the sociodemographic and clinical profiles of EOCRC vs. LOCRC hospitalizations using a nationally representative sample of 246,231 CRC hospitalizations from 2016 to 2020. Our analysis demonstrates meaningful differences in demographic characteristics, comorbidity burden, inpatient chemotherapy utilization, and inpatient mortality between EOCRC and LOCRC hospitalizations, fulfilling the primary and secondary aims of the study.

One of the most significant findings is the higher rate of EOCRC admissions among females and racial minorities, including Black, Hispanic, and Asian populations. This suggests potential disparities in risk factors or access to preventive measures among these groups. These differences underscore the need for targeted awareness and screening efforts, especially in underserved communities. These findings reaffirm prior studies which have shown the prevalence of EOCRC within African American and Hispanic communities, however it underscores the need for further research to be done among Asian/Pacific Islander population (14,15). The higher proportion of EOCRC admissions among younger females is particularly noteworthy, as it contrasts with the higher prevalence of LOCRC in males. This contrasts prior studies which show men having 16% higher rates of EOCRC (21). This difference may reflect patterns in hospitalizations between females and males with CRC which would need to be studied further. It is also interesting to note that the average age for EOCRC was 42.9 years, slightly lower than the current screening guidelines (3). These findings highlight the importance of early screening and risk stratification, especially in vulnerable communities.

The results also underscore the complex relationship between BMI and likelihood of CRC hospitalization. While ICD-10 coded obesity was associated with EOCRC hospitalizations, it was inversely related to LOCRC hospitalizations. This paradox may reflect differing biological mechanisms or lifestyle factors affecting CRC development across age groups (11). These findings should be interpreted with caution given the methodological limitations of BMI measurement in administrative databases. The high proportion of missing BMI data (75.7%) may limit the generalizability of these findings. Nevertheless, the observed association between obesity and early-onset CRC hospitalizations highlights a potential avenue for future prospective studies.

Interestingly, our study found that admissions for patients with EOCRC were more likely to have a history of tobacco use disorder and depression, whereas LOCRC admissions were associated with chronic medical conditions like diabetes and CAD. These findings suggest that psychosocial factors and lifestyle choices may play a more prominent role in the development of CRC in younger individuals (9), while metabolic and vascular health may be more critical in older adults (12).

The significantly higher rate of chemotherapy utilization among EOCRC (OR: 1.78) warrants careful interpretation. While this finding may reflect differences in disease biology, with EOCRC potentially presenting at more advanced stages requiring aggressive inpatient treatment, it may also reflect differences in treatment intensity, referral patterns or disease burden among younger patients. Inpatient chemotherapy administration may serve as a marker of disease severity, complications requiring hospitalization or differences in treatment strategies among younger patients. The decreased mortality risk and increased likelihood of receiving inpatient chemotherapy among EOCRC patients indicate potential differences in disease biology or treatment responses (16). Younger patients may present with more aggressive disease that prompts earlier and more intensive treatment, or they may have better overall health that enhances treatment tolerability and outcomes. Prior studies have shown that young adults with colon cancer received significantly more postoperative systemic chemotherapy at all stages, but they experienced only minimal gain in adjusted survival compared with their older counterparts who received less treatment (22). Notably, the NIS captures admission type, and future studies stratifying by elective vs. nonelective admissions within this population may help clarify whether inpatient chemotherapy utilization reflects planned treatment or response to acute complications such as bowel obstruction or perforation. Taken together, these findings underscore the complexity of interpreting inpatient chemotherapy utilization from administrative data alone and prospective studies are needed to better characterize treatment patterns and outcomes in EOCRC.

A recent meta-analysis on EOCRC vs. LOCRC concluded that despite the high heterogeneity of existing studies, EOCRC patients are diagnosed with significantly more advanced stages than LOCRC, although this is not reflected in any difference in cancer-related survival (23). Interestingly, the higher mortality risk in LOCRC admissions may be more complex and could be attributed to comorbidities and the decreased likelihood of receiving aggressive treatment. The lower inpatient mortality observed among EOCRC patients likely reflects differences in baseline health status and comorbidity burden compared with older patients. Younger patients may have fewer chronic medical conditions and may tolerate aggressive treatments more effectively, which could contribute to improved short-term inpatient outcomes despite potentially more advanced disease at presentation. Thus, the lower prevalence of chronic conditions such as diabetes and CAD in early-onset patients may also contribute to their better prognosis, as these comorbidities are known to complicate cancer treatment and recovery (24,25).

Our study also noted a lower prevalence of alcohol use disorder among EOCRC patients, which contrasts with the increased association in LOCRC. This could suggest different etiological pathways or risk profiles for alcohol consumption across different age groups. The increased association with alcohol use disorder in this group suggests that lifestyle modifications could be a critical component of prevention strategies for older adults (8). It also mirrors the Surgeon General’s advisory on alcohol consumption as a leading preventable cause of cancer (26).

Our study’s findings have several implications for clinical practice and public health policy. First, they validate the recent recommendations to lower the screening age for CRC to 45, particularly given the significant burden of EOCRC hospitalizations among younger adults (3). Additionally, the higher proportion of EOCRC hospitalizations among racial and ethnic minority populations and individuals with non-Medicare insurance highlights ongoing disparities in access to screening and early detection. Furthermore, the higher likelihood of inpatient chemotherapy among EOCRC patients may reflect differences in treatment patterns or disease severity at presentation. Further investigation into the drivers and outcomes of inpatient chemotherapy use in this population may help optimize treatment strategies and resource utilization. Our findings highlight the importance of considering sociodemographic factors and lifestyle behaviors in risk assessments and screening strategies. Tailored interventions that address the unique risk profiles of early-onset and LOCRC populations could improve prevention and treatment outcomes.

Limitations

This study has several limitations that must be considered when interpreting the results. First, the definitions for EOCRC and LOCRC are based on information available on the database with diagnosis of CRC. By filtering by age, there may be some patients with EOCRC who are misclassified as LOCRC. Second, the data used in this analysis is retrospective and derived from the NIS, which holds some limitations to the data available. The NIS is a hospitalization-level database and individual patients may contribute multiple hospitalizations across the study period, our results reflect hospitalizations rather than unique patients and findings cannot be used to estimate cancer incidence or population-level prevalence. Third, as a cross-sectional case-case comparison, this study captures a single snapshot of hospitalized CRC patients and cannot account for temporal dynamics or long-term carcinogenic processes that precede cancer detection. Factors present years to decades before diagnosis, including metabolic, environmental and behavioral exposures, may differently shape tumor characteristics and help explain the distinct clinical profiles observed between EOCRC and LOCRC and are not captured by this study design. Additionally, the NIS data does not include information on CRC subtypes or molecular markers, which could provide further insight into the mechanisms underlying the differences observed between early-onset and LOCRC.

Furthermore, EOCRC and LOCRC groups differ inherently by age, and many of the observed differences in comorbidity burden, inpatient mortality, and treatment patterns may reflect age-related factors rather than disease-specific characteristics of CRC onset. While multivariable regression was used to adjust for available covariates, residual confounding by age cannot be fully excluded and findings should be interpreted accordingly.

The assessment of obesity in this study carries important limitations. BMI data was missing in 75.7% of hospitalizations. This under documentation of obesity is known in the NIS where ICD-coded obesity prevalence has been shown to be significantly lower than prevalence based on measured BMI (27). BMI as a sole proxy for obesity may not fully capture the complexity of adiposity, as it does not account for factors such as body composition, fat distribution, or waist circumference. These factors collectively limit the strength of conclusions that can be drawn regarding the relationship between obesity and CRC hospitalization in this study.

While we analyzed a large sample size, our study is limited by the accuracy of diagnostic coding and the potential for misclassification of certain comorbid conditions. Future prospective studies incorporating more detailed clinical and molecular data would be valuable in addressing these limitations.

Conclusions

To our knowledge, this represents the first direct inpatient comparison of EOCRC and LOCRC across the full CRC spectrum using a nationally representative inpatient database. In conclusion, our analysis of population-level data from the NIS database adds to the growing body of evidence on the shifting landscape of CRC epidemiology, particularly in the context of early-onset disease. This nationally representative analysis of CRC hospitalizations from 2016–2020 demonstrates that EOCRC hospitalizations are characterized by a distinct inpatient profile relative to LOCRC, with marked racial and insurance disparities, higher rates of tobacco use disorder and depression, lower chronic comorbidity burden, and substantially higher inpatient chemotherapy utilization. These findings support the need for tailored screening guidelines and interventions that account for demographic, lifestyle, and behavioral health factors to effectively address the rising burden of EOCRC and improve outcomes for all patients. By addressing the specific needs of diverse patient populations, we can improve outcomes and reduce the burden of CRC across all age groups. Future research should focus on elucidating the underlying mechanisms driving these trends, exploring the impact of targeted prevention and treatment strategies, and leveraging methodologies such as the prospective cohort incident-tumor biobank method (PCIBM) to link long-term pre-diagnostic exposures with tumor molecular characteristics and better understand the biological and etiological drivers of EOCRC (28).

Acknowledgments

A preliminary version of this work was presented as an abstract at American College of Gastroenterology (ACG) Annual Meeting, October 2025.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2026-0285/rc

Peer Review File: Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2026-0285/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jgo.amegroups.com/article/view/10.21037/jgo-2026-0285/coif). S.L. reports advisory board participation and involvement in a clinical trial for AstraZeneca, both unrelated to colorectal cancer. The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. American Cancer Society. Colorectal Cancer Facts & Figures 2023-2025. Atlanta: American Cancer Society, 2023. Available online: https://www.cancer.org/content/dam/cancer-org/research/cancer-facts-and-statistics/colorectal-cancer-facts-and-figures/colorectal-cancer-facts-and-figures-2023.pdf
2. Siegel RL, Kratzer TB, Giaquinto AN, et al. Cancer statistics, 2025. CA Cancer J Clin 2025;75:10-45. [Crossref] [PubMed]
3. Wolf AMD, Fontham ETH, Church TR, et al. Colorectal cancer screening for average-risk adults: 2018 guideline update from the American Cancer Society. CA Cancer J Clin 2018;68:250-81. [Crossref] [PubMed]
4. Lawler T, Parlato L, Warren Andersen S. The histological and molecular characteristics of early-onset colorectal cancer: a systematic review and meta-analysis. Front Oncol 2024;14:1349572. [Crossref] [PubMed]
5. Pretta A, Ziranu P, Perissinotto E, et al. Early onset metastatic colorectal cancer patients as a distinctive clinical and molecular phenomenon. Br J Cancer 2025;132:188-94. [Crossref] [PubMed]
6. Bassett JK, Severi G, English DR, et al. Body size, weight change, and risk of colon cancer. Cancer Epidemiol Biomarkers Prev 2010;19:2978-86. [Crossref] [PubMed]
7. Valle L, Vilar E, Tavtigian SV, et al. Genetic predisposition to colorectal cancer: syndromes, genes, classification of genetic variants and implications for precision medicine. J Pathol 2019;247:574-88. [Crossref] [PubMed]
8. Aleksandrova K, Pischon T, Jenab M, et al. Combined impact of healthy lifestyle factors on colorectal cancer: a large European cohort study. BMC Med 2014;12:168. [Crossref] [PubMed]
9. Hofseth LJ, Hebert JR, Chanda A, et al. Early-onset colorectal cancer: initial clues and current views. Nat Rev Gastroenterol Hepatol 2020;17:352-64. [Crossref] [PubMed]
10. Renehan AG, Tyson M, Egger M, et al. Body-mass index and incidence of cancer: a systematic review and meta-analysis of prospective observational studies. Lancet 2008;371:569-78. [Crossref] [PubMed]
11. Bardou M, Barkun AN, Martel M. Obesity and colorectal cancer. Gut. 2013;62:933-47. [Crossref] [PubMed]
12. Aleksandrova K, Pischon T, Buijsse B, et al. Adult weight change and risk of colorectal cancer in the European Prospective Investigation into Cancer and Nutrition. Eur J Cancer 2013;49:3526-36. [Crossref] [PubMed]
13. Liu PH, Wu K, Ng K, et al. Association of Obesity With Risk of Early-Onset Colorectal Cancer Among Women. JAMA Oncol 2019;5:37-44. [Crossref] [PubMed]
14. Carethers JM. Clinical and Genetic Factors to Inform Reducing Colorectal Cancer Disparitites in African Americans. Front Oncol 2018;8:531. [Crossref] [PubMed]
15. Willauer AN, Liu Y, Pereira AAL, et al. Clinical and molecular characterization of early-onset colorectal cancer. Cancer 2019;125:2002-10. [Crossref] [PubMed]
16. Rogers JE, Johnson B. The reality of early-onset colorectal cancer: highlighting the needs in a unique but emerging population. Dig Med Res 2021;4:63. [Crossref] [PubMed]
17. Rogers JE, Woodard TL, Dasari A, et al. Fertility discussions in young adult stage III colorectal cancer population: a single-center institution experience. Support Care Cancer 2021;29:7351-4. [Crossref] [PubMed]
18. Suenghataiphorn T, Danpanichkul P, Kulthamrongsri N, et al. Disparity in trends and characteristics of early onset colorectal cancer: analysis from the National Inpatient Sample, 2016 to 2021. Proc (Bayl Univ Med Cent) 2024;37:928-33. [Crossref] [PubMed]
19. Healthcare Cost and Utilization Project (HCUP). HCUP National Inpatient Sample (NIS). Agency for Healthcare Research and Quality. 2012. [accessed June 2025]. Available online: www.hcup-us.ahrq.gov/nisoverview.jsp
20. StataCorp. Stata Statistical Software: Release 18. College Station (TX): StataCorp LLC; 2023.
21. Petrick JL, Barber LE, Warren Andersen S, et al. Racial Disparities and Sex Differences in Early- and Late-Onset Colorectal Cancer Incidence, 2001-2018. Front Oncol 2021;11:734998. [Crossref] [PubMed]
22. Kneuertz PJ, Chang GJ, Hu CY, et al. Overtreatment of young adults with colon cancer: more intense treatments with unmatched survival gains. JAMA Surg 2015;150:402-9. [Crossref] [PubMed]
23. Carbone F, Spinelli A, Ciardiello D, et al. Prognosis of early-onset versus late-onset sporadic colorectal cancer: Systematic review and meta-analysis. Eur J Cancer 2025;215:115172. [Crossref] [PubMed]
24. Michalopoulou E, Matthes KL, Karavasiloglou N, et al. Impact of comorbidities at diagnosis on the 10-year colorectal cancer net survival: A population-based study. Cancer Epidemiol 2021;73:101962. [Crossref] [PubMed]
25. Søgaard M, Thomsen RW, Bossen KS, et al. The impact of comorbidity on cancer survival: a review. Clin Epidemiol 2013;5:3-29. [Crossref] [PubMed]
26. US Department of Health and Human Services. Alcohol and Cancer Risk: The U.S. Surgeon General’s Advisory. 2025. Available online: https://www.hhs.gov/sites/default/files/oash-alcohol-cancer-risk.pdf
27. Arnold Z, Dang P, Hollman N, et al. Underdocumentation of Obesity in the National Inpatient Sample. Obesity (Silver Spring) 2026;34:329-33. [Crossref] [PubMed]
28. Ogino S, Ugai S, Hamada T, et al. The early-onset cancer epidemics: evidence synthesis using the prospective cohort incident-tumor biobank method. Eur J Epidemiol 2025;40:1405-17. [Crossref] [PubMed]