Abstract
We examined the relationships of sex and age specific groups with cause-specific survival for early-onset colorectal cancer (EOCRC) diagnosis. A retrospective cohort analysis utilizing data from the 2000-2020 Georgia Cancer Registry were performed. Sex and age at diagnosis were exposures of interest. CRC survival at 1-, 3, and 5-year intervals were our primary outcomes of interest. Traditional Cox proportional hazards regression and Piecewise Cox regression models were performed to examine the mentioned association. Among 11,935 EOCRC patients, males had lower 1- (89.4% vs. 91.9%), 3- (75.7% vs. 79.2%), and 5-year (69.7% vs. 74.3%) survival rates than female patients (all p-value <0.001). In adjusted analysis, regardless of survival intervals, male patients aged 30-39 years were more likely to die from CRC at 1-year (HR, 1.40; 95% CI, 1.08-1.82), 3-year (HR, 1.26; 95% CI, 1.06-1.49), 5-year (HR, 1.27; 95% CI, 1.09-1.48) than female aged 30-39 years, respectively. Our piecewise models also confirmed male patients aged 30-39 years were 33% more likely to die from CRC within 1 year interval. Similarly, male patients aged 40-49 years were more likely to die from CRC at 1-year (HR, 1.33; 95% CI, 1.16-1.53), 3-year (HR, 1.20; 95% CI, 1.10-1.32), and 5-year (HR, 1.22; 95% CI, 1.13-1.33) intervals than female patients aged 40-49 years, respectively. In summary, the highest estimate of EOCRC mortality within 1-year interval was observed among male patients aged 30-39 years. Prioritizing prevention and treatment strategies may reduce the risk of 1-year EOCRC mortality for males and 30-39 age group.
Keywords: Sex, colorectal cancer, survival intervals, age at diagnosis, young adults
Introduction
Early-onset colorectal cancer (EOCRC) has been reported to be increasing most rapidly in adults younger than 50 years old in the United States (US) [1]. An increase in mortality of EOCRC was observed by approximately 1.3% per year from 2008 to 2017 with the exception for individuals aged 20-29 years [2]. However, there is a lack of evidence-based interventions that effectively reduce EOCRC mortality. This is concerning because younger populations are currently under the recommended screening age and may misdiagnose due to limited healthcare access as a result of non-eligible for screening [3]. This is especially true for Georgia, where a large proportion of medically underserved communities is associated with higher EOCRC mortality rates (2.0 vs. 1.8 per 100,000 in the US) [4].
While the factors driving EOCRC mortality remain unclear, they likely involve both individual-level characteristics (e.g., sex, race, lifestyle factors, personal/family history of CRC) and area-level factors (e.g., access to health care services) [5-17]. Notably, research highlights age and sex as critical determinants of outcomes. Males demonstrate consistently higher mortality rates globally and, in the U.S., particularly in Southern US hotspots, with worse survival metrics than females [18-22]. Age-stratified analyses reveal distinct diagnostic patterns: advanced colon cancer predominates in patients aged 30-39 years versus advanced rectal cancer in those aged 20-29 years [23,24]. Therefore, the research examines integrated relationship of sex and age-related difference is critical because it can inform the targeted interventions aimed at prioritizing prevention strategies for high-risk young populations.
However, there is insufficient comprehensive research exploring cause-specific survival of EOCRC at different time intervals while considering sex and age-related differences simultaneously. To address the research gaps, we sought to examine the relationship of sex and age-related disparities with cause-specific survival at 1-, 3-, and 5-year intervals for EOCRC in Georgia by adjusting for important covariates (e.g., geographic factors related to Georgia). We also examined sex differences in 1-, 3-, and 5-year survival rates within three ages at diagnosis groups (18-29, 30-39, and 40-49 years) using 2000-2020 Georgia Cancer Registry. Several Southeastern states, such as South Carolina, share similar socioeconomic and demographic profiles with Georgia - including lower median household income and higher proportions of Black populations [25]. This more regionally focused approach may have greater implications for other states and regions in the US, particularly in the Southeast.
Methods
Study design
We conducted a retrospective study integrating patient-level and county-level information from publicly available databases and de-identified cancer registries. Patient-level data from the 2000-2020 Georgia Cancer Registry were requested through Georgia Department of Public Health, which is a source for comprehensive population-based information in Georgia that includes patient demographics, primary tumor site, tumor morphology and stage at diagnosis, and follow-up for vital status. Further, we gathered patient residential socioeconomic status using data from the U.S. Department of Agriculture (USDA) Economic Research Service (ERS), https://www.ers.usda.gov/. We linked USDA, ERS 2010 Rural-Urban Continuum (Beale) Code for rurality data and official ERS measure for persistent poverty data to patient-level data. The county Federal Information Processing System (FIPS) code was used as identifier to link Georgia Cancer Registry data (patient-level data) and geographic factors (county-level data) for CRC patients in Georgia. The study eligible population included patients diagnosed with CRC defined by the Surveillance, Epidemiology, and End Results (SEER) Site Recode ICD-O-3/WHO 2008 definition of colon cancer (C180-C189), rectosigmoid junction cancer (C199), and rectal cancer (C209). Data extracted for this study were approved by Institutional Review Boards (IRB) review at Georgia Department of Public Health, #240403.
Study participants
A total of 94,541 CRC patients were included between 2000 and 2020 in the Georgia Cancer Registry. We excluded 284 CRC patients aged under 18 years or with age unknown (n=108), missing information on race (n=112), missing survival time (n=1), CRC diagnosed before 2000 (n=27) and after 2021 (n=9), missing county code (n=8), unknown sex (n=6), and unknown primary site (n=13). The exclusion of cases diagnosed after 2021 due to limited sample size and because data beyond 2021 was not requested. To obtain an eligible study sample for EOCRC, we further excluded 82,322 patients who were diagnosed with CRC at the age of 50+ years. As a result, 11,935 patients aged 18-49 years diagnosed with CRC during 2000-2020 in Georgia were included as the study population for statistical analysis (Figure 1).
Figure 1.
Study sample selection. Abbreviations: CRC, colorectal cancer; EOCRC, early-onset colorectal cancer.
Measures
Sex (male or female at birth) and age at diagnosis (defined by around 10-year ranges within the definition of EOCRC, i.e., 18-29, 30-39, or 40-49 years) were exposures of interest. One-, three-, and five-year cause-specific survivals for CRC were our outcomes of interest. While the Five-year interval is a standard benchmark in cancer research, we also examined One- and Three-year survival intervals to highlight the urgency of EOCRC mortality among specific sex and age groups.
Covariates of interest include sociodemographic characteristics, geographic factors, tumor characteristics, and year of diagnosis. For sociodemographic characteristics, we included race [White, Black, Hispanic, Asian/Pacific Islander (API)/American Indian/Alaska Natives (AN/AI)], marital status (married, unmarried, unknown), and insurance status (public, private, and none/unknown). Unknown marital and insurance status were retained as distinct categories to improve sample size, rather than excluding these cases from analysis. For geographic factors, county-level rurality (no or yes), county-level persistent poverty (no or yes), and Georgia public health district (Southeast, Southwest, North Central, North, Northeast, Northwest, South, South Central, East Central, LaGrange, Coastal, West Central, North Georgia, Clayton, Cobb-Douglas, Fulton, DeKalb, East Metro). Persistent poverty was defined by using poverty rate 20% or higher in 1990, 2000, 2007-2011, and 2017-2021. For tumor characteristics, late stage at diagnosis (no or yes), high pathological grade (no or yes), and primary site (left or right) were included. Finally, we defined the years of diagnosis as 2000-2004, 2005-2009, 2010-2014, or 2015-2021.
Statistical analysis
Descriptive statistics (e.g., frequencies and relative frequencies) were used to describe age at diagnosis, sex, sociodemographic characteristics, tumor characteristics, and year of diagnosis. We also examined bivariate differences within age at diagnosis in sex, sociodemographic characteristics, geographic factors, tumor characteristics, and year of diagnosis, using chi-square tests. Patients’ survival time was measured in months from the date of diagnosis up to 12, 36, and 60 months of follow-up, censored at the date of last contact or death due to other causes. Survival analyses were applied using the Kaplan-Meier method with Log-rank test to compare the survival rates between males and females and stratified by three ages at diagnosis groups. Further, we performed traditional Cox proportional hazard regression to examine the association between sex, age at diagnosis, and causes-specific survival for CRC at 1-, 3-, and 5-year intervals. Unadjusted and adjusted models were performed to examine the association. Unadjusted models included sex and age at diagnosis only; adjusted models were further adjusted for sociodemographic characteristics, geographic factors, tumor characteristics, and year of diagnosis. Finally, we performed the stratify analyses to examine the relationship of sex with EOCRC mortality at three intervals across three age groups. All models were adjusted for sociodemographic characteristics, tumor characteristics, geographic factors, and year of diagnosis. To assess potential bias, sensitivity analyses were conducted by excluding cases with unknown marital and/or insurance status.
To elucidate the risk of EOCRC death in specific timeframe, we performed additional analyses by using piecewise Cox regression models with three follow-up time intervals (0-1 year, 1-5 years, 5+ years). The piecewise Cox model allows the hazard ratio (HR) to vary across different time intervals, which can overcome the limitation of traditional Cox model with constant HR assumption. All results were reported using HRs and the associated 95% confidence intervals (CIs). SAS Version 9.4, SAS Institute Inc., Cary, North Carolina was used to conduct all analyses. The level of statistical significance was set at an alpha level of 0.05 and the p-values were based on two-sided probability tests.
Results
In Table 1, the majority of EOCRC patients were aged 40-49 years (72.9%). Many were male (51.6%), were White (57.0%), were married (53.5%), had private insurance (44.6%), living in non-poverty (83.3%) or rural areas (81.5%), had no late stage at diagnosis (83.3%), had no high pathological grade (98.6%), had left-side CRC (68.5%), and were diagnosed during 2015-2020 (36.7%). LaGrange and Fulton districts (8.8% each) and Cobb-Douglas (8.5%) had the highest percentage of patients. Further, there were more female EOCRC patients diagnosed at age of 18-29 years while majority male patients diagnosed at age of 40-49 years (p-value =0.073).
Table 1.
Characteristics of colorectal cancer patients by age at diagnosis in Georgia (n=11,935)
| Total (n=11,935) | 18-29 (n=767, 6.4%) | 30-39 (n=2,469, 20.7%) | 40-49 (n=8,699, 72.9%) | P-value | |
|---|---|---|---|---|---|
| Sex | 0.073 | ||||
| Male | 6,155 (51.6%) | 376 (49.0%) | 1,239 (50.2%) | 4,540 (52.2%) | |
| Female | 5,780 (48.4%) | 391 (51.0%) | 1,230 (49.8%) | 4,159 (47.8%) | |
| Sociodemographic characteristics | |||||
| Race | <0.001 | ||||
| White | 6,798 (57.0%) | 442 (57.6%) | 1,391 (56.3%) | 4,965 (57.1%) | |
| Black | 4,150 (34.8%) | 236 (30.8%) | 827 (33.5%) | 3,087 (35.5%) | |
| Hispanic | 655 (5.5%) | 62 (8.1%) | 172 (7.0%) | 421 (4.8%) | |
| API/AN/AI | 332 (2.8%) | 27 (3.5%) | 79 (3.2%) | 226 (2.6%) | |
| Marital status | <0.001 | ||||
| Married | 6,383 (53.5%) | 176 (23.0%) | 1,326 (53.7%) | 4,881 (56.1%) | |
| Unmarried | 4,746 (39.8%) | 538 (70.1%) | 972 (39.4%) | 3,236 (37.2%) | |
| Unknown | 806 (6.8%) | 53 (6.9%) | 171 (6.9%) | 582 (6.7%) | |
| Insurance | <0.001 | ||||
| Public | 2,129 (17.8%) | 180 (23.5%) | 460 (18.6%) | 1,489 (17.1%) | |
| Private | 5,322 (44.6%) | 319 (41.6%) | 1,057 (42.8%) | 3,946 (45.4%) | |
| None/Unknown | 4,484 (37.6%) | 268 (34.9%) | 952 (38.6%) | 3,264 (37.5%) | |
| Geographic factors | |||||
| Poverty | 0.202 | ||||
| No | 9,937 (83.3%) | 711 (92.7%) | 2,309 (93.5%) | 8,043 (92.5%) | |
| Yes | 1,998 (16.7%) | 56 (7.3%) | 160 (6.5%) | 656 (7.5%) | |
| Rurality | 0.309 | ||||
| No | 9,730 (81.5%) | 633 (82.5%) | 2,034 (82.4%) | 7,063 (81.2%) | |
| Yes | 2,205 (18.5%) | 134 (17.5%) | 435 (17.6%) | 1,636 (18.8%) | |
| Public health district | 0.080 | ||||
| Southeast | 535 (4.5%) | 41 (5.4%) | 109 (4.4%) | 385 (4.4%) | |
| Southwest | 503 (4.2%) | 21 (2.7%) | 90 (3.7%) | 392 (4.5%) | |
| North Central | 633 (5.3%) | 48 (6.3%) | 120 (4.9%) | 465 (5.4%) | |
| North | 789 (6.6%) | 40 (5.2%) | 175 (7.1%) | 574 (6.6%) | |
| Northeast | 563 (4.7%) | 41 (5.4%) | 126 (5.1%) | 396 (4.6%) | |
| Northwest | 798 (6.7%) | 44 (5.7%) | 181 (5.1%) | 573 (6.6%) | |
| South | 314 (2.6%) | 20 (2.6%) | 71 (2.9%) | 223 (2.6%) | |
| South Central | 203 (1.7%) | 10 (1.3%) | 42 (1.7%) | 151 (1.7%) | |
| East Central | 574 (4.8%) | 29 (3.8%) | 113 (4.6%) | 432 (5.0%) | |
| LaGrange | 1,044 (8.8%) | 71 (9.3%) | 199 (8.1%) | 774 (8.9%) | |
| Coastal | 527 (4.4%) | 45 (5.9%) | 106 (4.3%) | 376 (4.3%) | |
| West Central | 437 (3.7%) | 23 (3.0%) | 80 (3.2%) | 334 (3.8%) | |
| North Georgia | 523 (4.4%) | 26 (3.4%) | 104 (4.2%) | 393 (4.5%) | |
| Clayton | 366 (3.1%) | 22 (2.9%) | 76 (3.1%) | 268 (3.1%) | |
| Cobb-Douglas | 1,013 (8.5%) | 61 (8.0%) | 201 (8.1%) | 751 (8.6%) | |
| Fulton | 1,047 (8.8%) | 88 (11.5%) | 226 (9.2%) | 733 (8.4%) | |
| DeKalb | 827 (6.9%) | 51 (6.7%) | 191 (7.7%) | 585 (6.7%) | |
| East Metro | 1,239 (0.4%) | 86 (11.2%) | 259 (10.5%) | 894 (10.3%) | |
| Tumor characteristics | |||||
| Late stage at diagnosis | 0.037 | ||||
| No | 9,937 (83.3%) | 664 (86.6%) | 2,044 (82.8%) | 7,229 (83.1%) | |
| Yes | 1,998 (16.7%) | 103 (13.4%) | 425 (17.2%) | 1,470 (16.9%) | |
| High pathological grade | 0.595 | ||||
| No | 11,770 (98.6%) | 758 (98.8%) | 2,430 (98.4%) | 8,582 (98.7%) | |
| Yes | 165 (1.4%) | 9 (1.2%) | 39 (1.6%) | 117 (1.3%) | |
| Primary site | <0.001 | ||||
| Left | 8,180 (68.5%) | 371 (48.4%) | 1,685 (68.3%) | 6,124 (70.4%) | |
| Right | 3,755 (31.5%) | 396 (51.6%) | 784 (31.8%) | 2,575 (29.6%) | |
| Year of diagnosis | <0.001 | ||||
| 2000-2004 | 2,330 (19.5%) | 104 (13.6%) | 507 (20.5%) | 1,719 (19.8%) | |
| 2005-2009 | 2,574 (21.6%) | 129 (16.8%) | 526 (21.3%) | 1,919 (22.1%) | |
| 2010-2014 | 2,656 (22.3%) | 154 (20.1%) | 507 (20.5%) | 1,995 (22.9%) | |
| 2015-2020 | 4,375 (36.7%) | 380 (37.6%) | 929 (37.6%) | 3,066 (35.3%) | |
Abbreviations: API, Asian/Pacific Islander; AI/AN, American Indian/Alaska Natives.
Survival analysis
Overall, male patients demonstrated the lower 1- (89.4% vs. 91.9%), 3- (75.7% vs. 79.2%), and 5-year (69.7% vs. 74.3%) survival time than female patients (all p-value <0.001). 5-year survival rates were lower among male patients aged 18-29 years (74.5% vs. 81.3% in female patients, p-value =0.029; Figure 2), aged 30-39 years (70.1% vs. 74.8% in female patients, p-value =0.004; Figure 3), and aged 40-49 years (69.2% vs. 73.6% in females, p-value <0.001; Figure 4). Male patients also demonstrated the lower survival rates at 1- and 3-year survival (Figures 2, 3 and 4).
Figure 2.
Kaplan-Meier curve of colorectal cancer survival at 18-29 age group.
Figure 3.
Kaplan-Meier curve of colorectal cancer survival at 30-39 age group.
Figure 4.
Kaplan-Meier curve of colorectal cancer survival at 40-49 age group.
In Table 2, results from unadjusted and adjusted models were similar for 5-year intervals. In the adjusted model, we observed that CRC patients aged 30-39 and 40-49 years were 28% (HR, 1.28; 95% CI, 1.08-1.52) and 33% (HR, 1.33; 95% CI, 1.13-1.56) more likely to die from CRC compared to patients aged 18-29, respectively. However, results from unadjusted and adjusted models were dissimilar for 3- and 1-year intervals. When adjusting for all covariates, we found that CRC patients aged 30-39 (HR, 1.21; 95% CI, 1.00-1.45) and 40-49 years were more likely to die from CRC at 3- (30-39 years: HR, 1.21; 95% CI, 1.00-1.45; 40-49 years: HR, 1.27; 95% CI, 1.07-1.51) and 1-year intervals (30-39 years: HR, 1.32; 95% CI, 1.00-1.76; 40-49 years: HR, 1.30; 95% CI, 1.00-1.69). Further, in full adjusted models, male patients demonstrated the greater risk of CRC death for 5-, 3-, and 1-year than female patients by 24% (HR, 1.24; 95% CI, 1.15-1.32), 22% (HR, 1.22; 95% CI, 1.13-1.31), and 34% (HR, 1.34; 95% CI, 1.19-1.31), respectively. The interaction between sex and age at diagnosis were not observed for three survival timeframes. Finally, our sensitivity analysis excluding unknown marital status and insurance status demonstrated similar results with full models (Supplementary Table 1).
Table 2.
Association between age at diagnosis, sex, and colorectal cancer survival time
| 1-year survival | 3-year survival | 5-year survival | ||||
|---|---|---|---|---|---|---|
|
|
|
|
||||
| Unadjusted model HR (95% CI) | Adjusted model HR (95% CI) | Unadjusted model HR (95% CI) | Adjusted model HR (95% CI) | Unadjusted model HR (95% CI) | Adjusted model HR (95% CI) | |
| Age at diagnosis | ||||||
| 18-29 years | Reference | Reference | Reference | Reference | Reference | Reference |
| 30-39 years | 1.19 (0.90, 1.58) | 1.32 (1.00, 1.76) | 1.12 (0.94, 1.35) | 1.21 (1.00, 1.45) | 1.21 (1.03, 1.44) | 1.28 (1.08, 1.52) |
| 40-49 years | 1.15 (0.89, 1.49) | 1.30 (1.00, 1.69) | 1.17 (0.99, 1.39) | 1.27 (1.07, 1.51) | 1.25 (1.14, 1.31) | 1.33 (1.13, 1.56) |
| Sex | ||||||
| Female | Reference | Reference | Reference | Reference | Reference | Reference |
| Male | 1.34 (1.19, 1.51) | 1.34 (1.19, 1.51) | 1.21 (1.12, 1.30) | 1.22 (1.13, 1.32) | 1.22 (1.14, 1.31) | 1.24 (1.16, 1.33) |
| P-value for interaction | 0.629 | 0.735 | 0.835 | 0.877 | 0.938 | 0.879 |
Abbreviations: HR, hazard ratio; CI, confidence interval. Italicized text indicates statistically significant results. Notes: Unadjusted model includes and age at diagnosis and sex only; Adjusted model was further adjusted for sociodemographic characteristics, geographic factors, tumor characteristics, and year of diagnosis.
As shown in Table 3, we examined the association between sex and 5-, 3-, and 1-year survival for CRC when stratified by three age groups. Regardless of survival intervals, male aged 30-39 years were more likely to die from CRC at 5- (HR, 1.27; 95% CI, 1.09-1.48), 3- (HR, 1.26; 95% CI, 1.06-1.49), and 1-year (HR, 1.40; 95% CI, 1.08-1.82) intervals, respectively. Similarly, male patients aged 40-49 years were more likely to die from CRC at 5- (HR, 1.22; 95% CI, 1.13-1.33), 3- (HR, 1.20; 95% CI, 1.10-1.32), and 1-year (HR, 1.33; 95% CI, 1.16-1.53) intervals than their female counterparts, respectively. Average survival months was also lower in male patients than female patients at 1-, 3-, and 5-year intervals for 30-39 and 40-49 age groups.
Table 3.
Association between sex and colorectal cancer survival time by age at diagnosis
| 1-year survival | 3-year survival | 5-year survival | ||||
|---|---|---|---|---|---|---|
|
|
|
|
||||
| Survival time Mean (SD) | HR (95% CI) | Survival time Mean (SD) | HR (95% CI) | Survival time Mean (SD) | HR (95% CI) | |
| 18-29 | ||||||
| Female | 11.0 (2.9) | Reference | 29.4 (11.4) | Reference | 43.1 (20.9) | Reference |
| Male | 10.9 (3.1) | 1.05 (0.61, 1.82) | 28.7 (11.9) | 1.15 (0.81, 1.64) | 42.1 (21.6) | 1.25 (0.90, 1.73) |
| 30-39 | ||||||
| Female | 11.3 (2.3) | Reference | 30.5 (10.3) | Reference | 45.5 (19.9) | Reference |
| Male | 11.1 (2.8) | 1.40 (1.08, 1.82) | 29.2 (11.4) | 1.26 (1.06, 1.49) | 43.0 (21.1) | 1.27 (1.09, 1.48) |
| 40-49 | ||||||
| Female | 11.3 (2.4) | Reference | 30.5 (10.3) | Reference | 45.4 (19.9) | Reference |
| Male | 11.1 (2.7) | 1.33 (1.16, 1.53) | 29.5 (11.1) | 1.20 (1.10, 1.32) | 43.6 (20.8) | 1.22 (1.13, 1.33) |
Abbreviations: HR, hazard ratio; CI, confidence interval. Italicized text indicates statistically significant results. Notes: All models were further adjusted for sociodemographic characteristics, geographic factors, tumor characteristics, and year of diagnosis.
In our piecewise Cox regression models, we observed that the risk of CRC death was 26% and 35% higher for those aged 30-39 and 40-49 between one to five years post diagnosis, respectively, when compared to patients aged 18-29 years, after adjusting for all covariates of interest (Table 4). Male patients were more likely to die from CRC regardless of survival intervals by 19%-34%. As shown in Table 5, we found that male patients aged 30-39 years had greater risk of CRC death within 1 year interval by 33% (HR, 1.33; 95% CI, 1.02-1.75). Regardless of survival interval, male patients aged 40-49 were also associated with increased risk of CRC death by 17%-38%.
Table 4.
Association between age at diagnosis, sex, and colorectal cancer survival time (0-1, 1-5, and 5+ Years)
| 0-1-year survival | 1-5-year survival | 5+-year survival | ||||
|---|---|---|---|---|---|---|
|
|
|
|
||||
| Unadjusted model HR (95% CI) | Adjusted model HR (95% CI) | Unadjusted model HR (95% CI) | Adjusted model HR (95% CI) | Unadjusted model HR (95% CI) | Adjusted model HR (95% CI) | |
| Age at diagnosis | ||||||
| 18-29 years | Reference | Reference | Reference | Reference | Reference | Reference |
| 30-39 years | 1.12 (0.84, 1.50) | 1.25 (0.94, 1.67) | 1.22 (0.99, 1.50) | 1.26 (1.02, 1.56) | 0.76 (0.51, 1.12) | 0.80 (0.54, 1.19) |
| 40-49 years | 1.08 (0.83, 1.41) | 1.23 (0.94, 1.61) | 1.30 (1.07, 1.58) | 1.35 (1.11, 1.64) | 1.01 (0.71, 1.43) | 1.11 (0.78, 1.58) |
| Sex | ||||||
| Female | Reference | Reference | Reference | Reference | Reference | Reference |
| Male | 1.32 (1.16, 1.49) | 1.32 (1.17, 1.50) | 1.17 (1.07, 1.27) | 1.19 (1.10, 1.29) | 1.31 (1.12, 1.53) | 1.34 (1.15, 1.57) |
Abbreviations: HR, hazard ratio; CI, confidence interval. Italicized text indicates statistically significant results. Notes: Unadjusted model includes and age at diagnosis and sex only; Adjusted model was further adjusted for sociodemographic characteristics, geographic factors, tumor characteristics, and year of diagnosis.
Table 5.
Association between sex and colorectal cancer survival time by age at diagnosis (0-1, 1-5, and 5+ Years)
| 0-1-year survival | 1-5-year survival | 5+-year survival | ||||
|---|---|---|---|---|---|---|
|
|
|
|
||||
| # of death (%) | HR (95% CI) | # of death (%) | HR (95% CI) | # of death (%) | HR (95% CI) | |
| 18-29 | ||||||
| Female | 28 (7.3%) | Reference | 46 (4.1%) | Reference | 16 (1.2%) | Reference |
| Male | 31 (8.5%) | 1.01 (0.58, 1.76) | 63 (6.1%) | 1.39 (0.93, 2.10) | 17 (1.3%) | 0.55 (0.21, 1.42) |
| 30-39 | ||||||
| Female | 96 (7.9%) | Reference | 211 (5.7%) | Reference | 51 (0.9%) | Reference |
| Male | 123 (10.2%) | 1.33 (1.02, 1.75) | 235 (6.7%) | 1.20 (0.99, 1.45) | 57 (1.1%) | 1.23 (0.84, 1.81) |
| 40-49 | ||||||
| Female | 308 (7.5%) | Reference | 773 (6.2%) | Reference | 218 (1.1%) | Reference |
| Male | 440 (9.9%) | 1.34 (1.15, 1.55) | 921 (7.1%) | 1.17 (1.06, 1.29) | 285 (1.5%) | 1.37 (1.15, 1.64) |
Abbreviations: HR, hazard ratio; CI, confidence interval. Italicized text indicates statistically significant results. Notes: All models were further adjusted for sociodemographic characteristics, geographic factors, tumor characteristics, and year of diagnosis.
Discussion
Our findings revealed that sex and age trend to be important factors on EOCRC mortality despite we adjusted for important covariates. We found that male patients aged 30-39 and 40-49 years were more likely to die from CRC by 20%-40% regardless of survival intervals. A significant higher estimate of EOCRC mortality at 1-year interval was observed for males aged 30-39 years. Such evidence highlights the need for prioritizing prevention and treatment strategies for males aged 30-39 years in Georgia.
Overall, we found that older age was associated with greater risk of death from CRC regardless of survival intervals by 21%-33%. Similar to our findings, a recent study reported that older patients (30-39 and 40-49 age groups) were more likely to die from non-cancer causes [18]. However, this study did not find the significant difference for cause-specific death for EOCRC [18]. This discrepancy may be explained by the different geographic populations. In particular, our population is from a single state (Georgia) with a large number of underserved communities that have been considered as hotspots of EOCRC mortality [21]. Several regions in Georgia are considered as medically underserved areas [26]. Living in socioeconomically disadvantaged neighborhoods or medically underserved areas were also more likely to die from CRC death by 15-25% [27,28]. More importantly, the impact of socioeconomically disadvantaged characteristics (e.g., living in rural or persistent poverty areas) was more pronounced for EOCRC by 30% [24]. Despite the evidence, in our multivariable analysis, we did not find the significant impact of rurality or poverty on CRC mortality regardless of time intervals. Majority of young patients tend to live in non-rural or poverty areas (Table 1). Future studies explore the other barriers (e.g., lack of awareness of CRC risk) may further elucidate this phenomenon.
Further, we found that male patients were associated with increased risk of CRC death regardless of survival intervals by 22%-34% compared to females. The survival rates were also lower for males than females overall and for males aged 30-39 and 40-49 years. In particular, findings from our piecewise Cox models further confirmed the CRC death fallen in 1-5 years interval for 30-49 age groups and male patients (Table 4). In line with prior literature, Afify and colleagues reported that the 1-year, 3-year, and 5-year overall survival rates for young males (89.1%, 73.8%, 65.7%) were lower than young females (91.3%, 76.9%, 69.9%) [18]. In the same study, authors also reported that males were associated with worse casus-specific survival for EOCRC by 15% [18]. However, a study from UK revealed that males had slightly better 1-year survival than females, but the 5-year survival appeared similar between males and females [29]. The discrepancy between our findings and those of the UK study may be attributed to the latter’s lack of adjustment for key covariates - such as sociodemographic factors, tumor characteristics, and geographic variables - in multivariable analysis, which are known to influence CRC survival outcomes. Additionally, variations in patient demographics, socioeconomic factors, and healthcare systems are likely to contribute to these differences, underscoring the need to examine specific populations to better understand the factors influencing health outcomes. Despite our findings being consistent with prior literature in general, none of these studies considered both sex and age at diagnosis for EOCRC survival. Thus, our findings can add to the existing evidence by specifying sex and age population.
In stratified analysis, we found the higher estimates of EOCRC death was observed in male patients aged 30-39 years at 1-year interval by 41%, compared to female patients aged 30-39 years. This phenomenon also confirmed by our piecewise Cox models with 33% increased risk of EOCRC death at 0-1-year interval for male patients aged 30-39 years (Table 5). Since no studies have examined EOCRC mortality across different time intervals while simultaneously accounting for sex and age-related differences, direct comparisons with previous literature are not feasible. Yet, by using SEER program, Afify and colleagues reported that male aged 30-39 and 40-49 years were 1.8-3.2-fold more likely to die from non-cancer related causes [18]. Several factors such as molecular and genetic factors and lifestyle may be attributed to the unfavorable survival in males [30]. For example, males usually have higher smoking [31] and heavier alcohol consumption rates compared to females [32]. Males are also more likely to consume meat, red meat, and processed meat than females [33]. They also tend to have higher relative weight or body mass index (BMI), which is considered a potential risk factor for CRC [33]. All these risk factors associated with worse health behaviors in males may explain the poorer causes-specific survival for EOCRC observed in our study, particularly at 1-year interval in our analysis. This is also particular true for Georgia because nearly half of cancer deaths in Georgia are linked to modifiable behavioral risk factors such as obesity (BMI ≥30) and smoking [34-38]. Such evidence highlights the need for examining various factors (e.g., health behaviors) associated with these pheromones in Georgia, particularly for males aged 30-39 years.
The main strength of this research is the first to investigate sex and age-related differences in EOCRC survival while considering three survival intervals for EOCRC by using traditional and piecewise Cox regression models. By using two different approaches, we are able to confirm the higher risk of EOCRC death in male patients aged 30-39 years within 1 year interval. This finding has important implications for this age group. Tailored education programs to raise awareness of CRC risk, along with targeted prevention strategies such as risk assessment or shortening follow-up time, may be beneficial. More research should also further investigate boarder factors such as biological, behavioral, or healthcare access contributing to early mortality in this group. Despite the strengths, there were a few limitations that should be noted. Although the study has accounted for many covariates in analyses, the Georgia cancer registry database did not collect information on individual income, education level, healthy lifestyle factors (e.g., body mass index, smoking, and alcohol consumption), and concomitant diseases [39,40]. For example, obesity is strongly associated with EOCRC risk and may lead to the greater risk of CRC-related death [41-43]. Future research should integrate multiple data sources to capture relevant factors and further clarify the underlying survival disparities. Second, although our study population is from Georgia, which has similar demographic characteristics in the Southeast states [25], generalizability issue may still exist. However, our study highlights the value of focusing on more narrowly defined populations that may be affected by similar socioeconomic factors, healthcare systems and other demographic characteristics (e.g., racial minorities). This approach can serve as a foundation for studies involving different geographic locations, including Southeast states (e.g., South Carolina), particularly given that 92% of counties in South have been identified as EOCRC hotspots [21]. Further, we were unable to include potential barriers to care (e.g., lack of awareness of cancer risk, lack of healthcare access for treatment resources) that may partly explain our findings regarding the higher EOCRC mortality for male patients and those aged 30-49 years. We also did not include factors associated with cancer prevention, such as family history of CRC. Finally, the retrospective nature of the study relies on existing data, which may be subject to biases such as incomplete records or inaccuracies in coding.
Conclusion
Age related disparities in EOCRC mortality were observed among male patients. Male patients aged 30-49 years demonstrated the greater risk of EOCRC death at 1-, 3- and 5- year intervals. In particular, the greater risk of EOCRC mortality at 1-year interval was observed among males aged 30-39 years. Prioritizing prevention and treatment strategies may reduce the risk of 1-year EOCRC mortality for males and 30-39 age group. Improving awareness of CRC risk and promoting healthy lifestyle for these groups may also be beneficial. Future research should evaluate multifaceted factors and identify/address specific barriers among young males to reduce disparities in EOCRC mortality in Georgia.
Acknowledgements
Dr. Meng-Han Tsai was supported by at least in part through the Georgia Cancer Center Paceline funding mechanism at Augusta University (MCGFD01071), National Cancer Institute (NCI, R21CA301113), and Augusta ROAR SCORE Career Enhancement Core (U54HL169191). Dr. Justin Moore was supported through National Cancer Institute (R21CA301113). The content is solely the responsibility of the authors and does not necessarily represent the official views of Augusta University and NCI.
Disclosure of conflict of interest
None.
Supporting Information
References
- 1.Shah RR, Millien VO, da Costa WL Jr, Oluyomi AO, Gould Suarez M, Thrift AP. Trends in the incidence of early-onset colorectal cancer in all 50 United States from 2001 through 2017. Cancer. 2022;128:299–310. doi: 10.1002/cncr.33916. [DOI] [PubMed] [Google Scholar]
- 2.Siegel RL, Miller KD, Goding Sauer A, Fedewa SA, Butterly LF, Anderson JC, Cercek A, Smith RA, Jemal A. Colorectal cancer statistics, 2020. CA Cancer J Clin. 2020;70:145–164. doi: 10.3322/caac.21601. [DOI] [PubMed] [Google Scholar]
- 3.Done JZ, Fang SH. Young-onset colorectal cancer: a review. World J Gastrointest Oncol. 2021;13:856–866. doi: 10.4251/wjgo.v13.i8.856. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.National Cancer Institute. Death Rate Report for Georgia by County, Colon & Rectum, 2016-2020. Accessed March 14, 2023. [Google Scholar]
- 5.Ahnen DJ, Wade SW, Jones WF, Sifri R, Mendoza Silveiras J, Greenamyer J, Guiffre S, Axilbund J, Spiegel A, You YN. The increasing incidence of young-onset colorectal cancer: a call to action. Mayo Clin Proc. 2014;89:216–224. doi: 10.1016/j.mayocp.2013.09.006. [DOI] [PubMed] [Google Scholar]
- 6.Chen H, Zheng X, Zong X, Li Z, Li N, Hur J, Fritz CD, Chapman W Jr, Nickel KB, Tipping A, Colditz GA, Giovannucci EL, Olsen MA, Fields RC, Cao Y. Metabolic syndrome, metabolic comorbid conditions and risk of early-onset colorectal cancer. Gut. 2021;70:1147–1154. doi: 10.1136/gutjnl-2020-321661. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Daca-Alvarez M, Marti M, Spinelli A, de Miranda NFFC, Palles C, Vivas A, Lachtford A, Monahan K, Szczepkowski M, Tarnowski W, Makkai-Popa ST, Vidal R, Lopez I, Hurtado E, Jimenez F, Jimenez-Toscano M, Alvaro E, Sanz G, Ballestero A, Melone S, Brandariz L, Prieto I, Garcia-Olmo D, Ocana T, Moreira R, Moreno L, Carballal S, Moreira L, Pellise M, Gonzalez-Sarmiento R, Holowatyj AN, Perea J, Balaguer F GEOCODE and SECOC consortia. Familial component of early-onset colorectal cancer: opportunity for prevention. Br J Surg. 2022;109:1319–1325. doi: 10.1093/bjs/znac322. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Danial D, Youssef ED, Maryam BM, Mohammad A, Moein BM, Liliane D. Risk factors of young-onset colorectal cancer: analysis of a large population-based registry. Can J Gastroenterol Hepatol. 2022;2022:3582443. doi: 10.1155/2022/3582443. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Deng JW, Zhou YL, Dai WX, Chen HM, Zhou CB, Zhu CQ, Ma XY, Pan SY, Cui Y, Xu J, Zhao EH, Wang M, Chen JX, Wang Z, Liu Q, Wang JL, Cai GX, Chen YX, Fang JY. Noninvasive predictive models based on lifestyle analysis and risk factors for early-onset colorectal cancer. J Gastroenterol Hepatol. 2023;38:1768–1777. doi: 10.1111/jgh.16243. [DOI] [PubMed] [Google Scholar]
- 10.Gausman V, Dornblaser D, Anand S, Hayes RB, O’Connell K, Du M, Liang PS. Risk factors associated with early-onset colorectal cancer. Clin Gastroenterol Hepatol. 2020;18:2752–2759. e2. doi: 10.1016/j.cgh.2019.10.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Hur J, Otegbeye E, Joh HK, Nimptsch K, Ng K, Ogino S, Meyerhardt JA, Chan AT, Willett WC, Wu K, Giovannucci E, Cao Y. Sugar-sweetened beverage intake in adulthood and adolescence and risk of early-onset colorectal cancer among women. Gut. 2021;70:2330–2336. doi: 10.1136/gutjnl-2020-323450. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Lamprell K, Pulido DF, Arnolda G, Easpaig BNG, Tran Y, Owais SS, Liauw W, Braithwaite J. People with early-onset colorectal cancer describe primary care barriers to timely diagnosis: a mixed-methods study of web-based patient reports in the United Kingdom, Australia and New Zealand. BMC Prim Care. 2023;24:12. doi: 10.1186/s12875-023-01967-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Li H, Chen X, Hoffmeister M, Brenner H. Associations of smoking with early- and late-onset colorectal cancer. JNCI Cancer Spectr. 2023;7:pkad004. doi: 10.1093/jncics/pkad004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Low EE, Demb J, Liu L, Earles A, Bustamante R, Williams CD, Provenzale D, Kaltenbach T, Gawron AJ, Martinez ME, Gupta S. Risk factors for early-onset colorectal cancer. Gastroenterology. 2020;159:492–501. e7. doi: 10.1053/j.gastro.2020.01.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Ochs-Balcom HM, Kanth P, Cannon-Albright LA. Early-onset colorectal cancer risk extends to second- and third-degree relatives. Cancer Epidemiol. 2021;73:101973. doi: 10.1016/j.canep.2021.101973. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Tiritilli A, Ko C. Patients with early-onset colorectal cancer have an increased risk of second primary malignancy. Dig Dis Sci. 2022;67:1328–1336. doi: 10.1007/s10620-021-06971-x. [DOI] [PubMed] [Google Scholar]
- 17.Zheng X, Hur J, Nguyen LH, Liu J, Song M, Wu K, Smith-Warner SA, Ogino S, Willett WC, Chan AT, Giovannucci E, Cao Y. Comprehensive assessment of diet quality and risk of precursors of early-onset colorectal cancer. J Natl Cancer Inst. 2021;113:543–552. doi: 10.1093/jnci/djaa164. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Afify AY, Ashry MH, Hassan H. Sex differences in survival outcomes of early-onset colorectal cancer. Sci Rep. 2024;14:22041. doi: 10.1038/s41598-024-71999-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Crosbie AB, Roche LM, Johnson LM, Pawlish KS, Paddock LE, Stroup AM. Trends in colorectal cancer incidence among younger adults-Disparities by age, sex, race, ethnicity, and subsite. Cancer Med. 2018;7:4077–4086. doi: 10.1002/cam4.1621. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Pan H, Zhao Z, Deng Y, Zheng Z, Huang Y, Huang S, Chi P. The global, regional, and national early-onset colorectal cancer burden and trends from 1990 to 2019: results from the Global Burden of Disease Study 2019. BMC Public Health. 2022;22:1896. doi: 10.1186/s12889-022-14274-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Rogers CR, Moore JX, Qeadan F, Gu LY, Huntington MS, Holowatyj AN. Examining factors underlying geographic disparities in early-onset colorectal cancer survival among men in the United States. Am J Cancer Res. 2020;10:1592–1607. [PMC free article] [PubMed] [Google Scholar]
- 22.Sheneman DW, Finch JL, Messersmith WA, Leong S, Goodman KA, Davis SL, Purcell WT, McCarter M, Gajdos C, Vogel J, Eckhardt SG, Lieu CH. The impact of young adult colorectal cancer: incidence and trends in Colorado. Colorectal Cancer. 2017;6:49–56. [Google Scholar]
- 23.Montminy EM, Zhou M, Maniscalco L, Penrose H, Yen T, Patel SG, Wu XC, Karlitz JJ. Trends in the incidence of early-onset colorectal adenocarcinoma among Black and White US residents aged 40 to 49 years, 2000-2017. JAMA Netw Open. 2021;4:e2130433. doi: 10.1001/jamanetworkopen.2021.30433. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Tsai MH, Coughlin SS, Cortes J, Thompson CA. Intersection of poverty and rurality for early-onset colorectal cancer survival. JAMA Netw Open. 2024;7:e2430615. doi: 10.1001/jamanetworkopen.2024.30615. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.State Cancer Profiles- Quick Profiles: South Carolina. Accessed August 4, 2025. [Google Scholar]
- 26.Brewer R, Goble G, Guy P. A peach of a telehealth program: Georgia connects rural communities to better healthcare. Perspect Health Inf Manag. 2011;8:1c. [PMC free article] [PubMed] [Google Scholar]
- 27.Singh GK, Williams SD, Siahpush M, Mulhollen A. Socioeconomic, rural-urban, and racial inequalities in US cancer mortality: part I-all cancers and lung cancer and part II-colorectal, prostate, breast, and cervical cancers. J Cancer Epidemiol. 2011;2011:107497. doi: 10.1155/2011/107497. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Hines RB, Markossian TW. Differences in late-stage diagnosis, treatment, and colorectal cancer-related death between rural and urban African Americans and Whites in Georgia. J Rural Health. 2012;28:296–305. doi: 10.1111/j.1748-0361.2011.00390.x. [DOI] [PubMed] [Google Scholar]
- 29.White A, Ironmonger L, Steele RJC, Ormiston-Smith N, Crawford C, Seims A. A review of sex-related differences in colorectal cancer incidence, screening uptake, routes to diagnosis, cancer stage and survival in the UK. BMC Cancer. 2018;18:906. doi: 10.1186/s12885-018-4786-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Fernandez E, Bosetti C, La Vecchia C, Levi F, Fioretti F, Negri E. Sex differences in colorectal cancer mortality in Europe, 1955-1996. Eur J Cancer Prev. 2000;9:99–104. doi: 10.1097/00008469-200004000-00005. [DOI] [PubMed] [Google Scholar]
- 31.Chang LC, Wu MS, Tu CH, Lee YC, Shun CT, Chiu HM. Metabolic syndrome and smoking may justify earlier colorectal cancer screening in men. Gastrointest Endosc. 2014;79:961–969. doi: 10.1016/j.gie.2013.11.035. [DOI] [PubMed] [Google Scholar]
- 32.Schutze M, Boeing H, Pischon T, Rehm J, Kehoe T, Gmel G, Olsen A, Tjonneland AM, Dahm CC, Overvad K, Clavel-Chapelon F, Boutron-Ruault MC, Trichopoulou A, Benetou V, Zylis D, Kaaks R, Rohrmann S, Palli D, Berrino F, Tumino R, Vineis P, Rodriguez L, Agudo A, Sanchez MJ, Dorronsoro M, Chirlaque MD, Barricarte A, Peeters PH, van Gils CH, Khaw KT, Wareham N, Allen NE, Key TJ, Boffetta P, Slimani N, Jenab M, Romaguera D, Wark PA, Riboli E, Bergmann MM. Alcohol attributable burden of incidence of cancer in eight European countries based on results from prospective cohort study. BMJ. 2011;342:d1584. doi: 10.1136/bmj.d1584. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Feraco A, Gorini S, Camajani E, Filardi T, Karav S, Cava E, Strollo R, Padua E, Caprio M, Armani A, Lombardo M. Gender differences in dietary patterns and physical activity: an insight with principal component analysis (PCA) J Transl Med. 2024;22:1112. doi: 10.1186/s12967-024-05965-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Alwers E, Carr PR, Banbury B, Walter V, Chang-Claude J, Jansen L, Drew DA, Giovannucci E, Nan H, Berndt SI, Huang WY, Prizment A, Hayes RB, Sakoda LC, White E, Labadie J, Slattery M, Schoen RE, Diergaarde B, van Guelpen B, Campbell PT, Peters U, Chan AT, Newcomb PA, Hoffmeister M, Brenner H. Smoking behavior and prognosis after colorectal cancer diagnosis: a pooled analysis of 11 studies. JNCI Cancer Spectr. 2021;5:pkab077. doi: 10.1093/jncics/pkab077. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Avgerinos KI, Spyrou N, Mantzoros CS, Dalamaga M. Obesity and cancer risk: emerging biological mechanisms and perspectives. Metabolism. 2019;92:121–135. doi: 10.1016/j.metabol.2018.11.001. [DOI] [PubMed] [Google Scholar]
- 36.Centers for Disease Control and Prevention. BRFSS Prevalence & Trends Data [online] Accessed August 22, 2024. [Google Scholar]
- 37.County Health Rankings. Georgia. Accessed September 4, 2024. [Google Scholar]
- 38.Izano M, Satariano WA, Hiatt RA, Braithwaite D. Smoking and mortality after breast cancer diagnosis: the health and functioning in women study. Cancer Med. 2015;4:315–324. doi: 10.1002/cam4.359. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Coughlin SS. Social determinants of colorectal cancer risk, stage, and survival: a systematic review. Int J Colorectal Dis. 2020;35:985–995. doi: 10.1007/s00384-020-03585-z. [DOI] [PubMed] [Google Scholar]
- 40.van Zutphen M, Kampman E, Giovannucci EL, van Duijnhoven FJB. Lifestyle after colorectal cancer diagnosis in relation to survival and recurrence: a review of the literature. Curr Colorectal Cancer Rep. 2017;13:370–401. doi: 10.1007/s11888-017-0386-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Xu P, Tao Z, Yang H, Zhang C. Obesity and early-onset colorectal cancer risk: emerging clinical evidence and biological mechanisms. Front Oncol. 2024;14:1366544. doi: 10.3389/fonc.2024.1366544. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Li H, Boakye D, Chen X, Hoffmeister M, Brenner H. Association of body mass index with risk of early-onset colorectal cancer: systematic review and meta-analysis. Am J Gastroenterol. 2021;116:2173–2183. doi: 10.14309/ajg.0000000000001393. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Paragomi P, Zhang Z, Abe SK, Islam MR, Rahman MS, Saito E, Shu XO, Dabo B, Pham YT, Chen Y, Gao YT, Koh WP, Sawada N, Malekzadeh R, Sakata R, Hozawa A, Kim J, Kanemura S, Nagata C, You SL, Ito H, Park SK, Yuan JM, Pan WH, Wen W, Wang R, Cai H, Tsugane S, Pourshams A, Sugawara Y, Wada K, Chen CJ, Oze I, Shin A, Ahsan H, Boffetta P, Chia KS, Matsuo K, Qiao YL, Rothman N, Zheng W, Inoue M, Kang D, Luu HN. Body mass index and risk of colorectal cancer incidence and mortality in Asia. JAMA Netw Open. 2024;7:e2429494. doi: 10.1001/jamanetworkopen.2024.29494. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.