Role of Demographics in Non-Invasive Testing for Colorectal Cancer Screening: Do Targeted Cut-Off Values Improve Detection?

Inayat Gill, Beaumont Resident
Christienne Shams, Beaumont Health Resident
Angy Hanna, Beaumont Health Fellow
Julie George, Beaumont Health
Laith H. Jamil, Beaumont Health
Atulkumar Patel, Beaumont Health


Introduction: Colorectal cancer (CRC) is the third leading cause of cancer-related deaths in the United States. For those unable or unwilling to undergo invasive screening, fecal immunochemical test (FIT) is a yearly alternative due to its cost and accessibility compared to the Cologuard. Few studies have examined how patient factors influence FIT outcomes. This study aims to determine whether FIT should be interpreted within the context of patient demographics and medical history. Methods: A retrospective review of patients .50 years old who had a FIT followed by a colonoscopy within 1 year from 2016-2019 was performed. Patients were analyzed based on age, race, body mass index (BMI), comorbidities, smoking, and alcohol use. FIT values above 100 ng/mL were positive based on manufacturer recommendations (Polymedco, NY). Chi-square test was used to compare positive FIT rates within each of the patient demographics and medical history. Fisher’s exact test was used when sparse cells were encountered in contingency tables. False positive (FP) and negative (FN) FIT results were determined by comparing with the gold standard of colonoscopy. Results: 1025 patients were reviewed. 21.8% of FIT results were positive. Factors which differed in positive FIT rates were age (P 5 0.003), smoking (P , 0.001), alcohol (P 5 0.001), and hypertension (P , 0.001). A significant trend was noted with increasing positive findings with older age (CochranArmitage test, P 5 0.001). The difference in rates of FP and FN FIT outcomes among each variable underwent further sub-analysis. The FP rate was 66.8% and the FN rate was 12.8%. Higher FN outcomes were noted in those above the age of 70, males and smokers, though the result was only statistically significant for males (P 5 0.009). Females were observed to have higher FP rates than males (P 5 0.019). African Americans had the highest rates of a positive FIT overall, and the FN rates were the lowest; neither result was statistically significant. Patients on aspirin and with a history of alcohol use had higher FP rates but did not meet statistical significance. Conclusion: Females were found to have higher FP FIT rates compared to males indicating that patient demographics may influence the FIT outcome. We can use this information to identify populations at higher risk of FP or FN FIT outcomes to target CRC screening. Additionally, recalculating the FP and FN rates for each variable by adjusting FIT cutoff may help determine new FIT targets to better improve sensitivity and specificity.