Type 2 Diabetes and Colorectal Cancer Risk

Abstract

Importance: Type 2 diabetes and colorectal cancer (CRC) disproportionately burden indviduals of low socioeconomic status and African American race. Although diabetes is an emerging CRC risk factor, associations between diabetes and CRC in these populations are understudied.

Objective: To determine if diabetes is associated with CRC risk in a cohort representing understudied populations.

Design, setting, and participants: This cohort study uses data from the prospective Southern Community Cohort Study in the US, which recruited from 2002 to 2009 and completed 3 follow-up surveys by 2018. Of about 85 000 participants, 86% enrolled at community health centers, while 14% were enrolled via mail or telephone from the same 12 recruitment states. Participants with less than 2 years of follow-up, previous cancer diagnosis (excluding nonmelanoma skin cancer) at enrollment, missing enrollment diabetes status, diabetes diagnosis before age 30, and without diabetes at enrollment with no follow-up participation were excluded. Data were analyzed from January to September 2023.

Exposures: Physician-diagnosed diabetes and age at diabetes diagnosis were self-reported via survey at enrollment and 3 follow-ups.

Main outcomes and measures: Diabetes diagnosis was hypothesized to be positively associated with CRC risk before analysis. Incident CRC was assessed via state cancer registry and National Death Index linkage. Hazard ratios and 95% CIs were obtained via Cox proportional hazard models, using time-varying diabetes exposure.

Results: Among 54 597 participants, the median (IQR) enrollment age was 51 (46-58) years, 34 786 (64%) were female, 36 170 (66%) were African American, and 28 792 (53%) had income less than $15 000 per year. In total, 289 of 25 992 participants with diabetes developed CRC, vs 197 of 28 605 participants without diabetes. Diabetes was associated with increased CRC risk (hazard ratio [HR], 1.47; 95% CI, 1.21-1.79). Greater associations were observed among participants without colonoscopy screening (HR, 2.07; 95% CI, 1.16-3.67) and with smoking history (HR, 1.62; 95% CI, 1.14-2.31), potentially due to cancer screening differences. Greater associations were also observed for participants with recent diabetes diagnoses (diabetes duration <5 years compared with 5-10 years; HR, 2.55; 95% CI, 1.77-3.67), possibly due to recent screening.

Conclusions and relevance: In this study where the majority of participants were African American with low socioeconomic status, diabetes was associated with elevated CRC risk, suggesting that diabetes prevention and control may reduce CRC disparities. The association was attenuated for those who completed colonoscopies, highlighting how adverse effects of diabetes-related metabolic dysregulation may be disrupted by preventative screening.

Figure 1.. Associations Between Diabetes and Incident Colorectal Cancer (CRC), Stratified by Risk Factors for CRC^a,b

Data presented as hazard ratios (HRs; with 95% CIs) for incident CRC in persons with diabetes compared with persons without diabetes. Diabetes includes prevalent diabetes at enrollment and incident diabetes during follow-up. BMI indicates body mass index (calculated as weight in kilograms divided by height in meters squared); CRC, colorectal cancer. ^aParticipants without diabetes who did not attend follow-up 3 were censored at their age at the last follow-up where they participated. ^bAnalyses adjusted for enrollment source, race and ethnicity, sex, health insurance status, CRC screening at enrollment, smoking status, education, income, alcohol intake, BMI (calculated as weight in kilograms divided by height in meters squared), and family history of colorectal cancer.

Figure 2.. Association Between Diabetes and Risk for Colorectal Cancer (CRC) Stratified by Colonoscopy Screening Status

Data presented as hazard ratios (HRs; with 95% CIs) for incident CRC in persons with diabetes compared with persons without diabetes. Analyses adjusted for enrollment source, race and ethnicity, sex, health insurance status, CRC screening at enrollment, smoking status, education, income, alcohol intake, body mass index, and family history of colorectal cancer. HR indicates hazard ratio.