Demographic, clinical, and outcomes characteristics associated with screening colonoscopy in colorectal cancer patients.

Abstract

554 Background: Screening colonoscopy has well established role in CRC prevention. Factors associated with underutilization of screening colonoscopy, and how underutilization affects the CRC outcomes is unclear. We conducted a retrospective study with an aim to identify demographic, clinical and outcome characteristics associated with screening colonoscopy in patients with CRC. Methods: The Indiana Network for Patient Care (INPC) was used to identify patients diagnosed with CRC between 2001-2015. The INPC is the largest and longest tenured clinical data warehouse of the Indiana Health Information Exchange in the USA. Patient demographic (age, race, gender, median household income, insurance type, geographical location), clinical (BMI, year of diagnosis, stage, tumor location, CEA, surgery, chemotherapy, comorbidity) and outcome characteristics were obtained. Only patients who had colonoscopy prior to diagnosis were included, and divided those who received screening colonoscopy versus diagnostic colonoscopy. Chi-square was used for univariate analysis. Multivariate logistic regression was used to model the association of colonoscopy with overall mortality, and CRC-specific mortality. Results: A total of 1546 patients were identified, of which 361 (23.3%) and 1185 (76.6%) had screening and diagnostic colonoscopy, respectively. On univariate analyses, older age, female gender, Caucasian race, lower BMI and lack of insurance, were significantly associated with screening colonoscopy (P < 0.05). Additionally, patients receiving screening colonoscopy had higher likelihood to undergo surgery and receive chemotherapy (P < 0.05). Multivariate analyses adjusted for age, gender, race, surgery, chemotherapy and insurance status showed that screening colonoscopy (as compared with diagnostic colonoscopy) is associated with a 38% lower odds of overall mortality (adjusted OR = 0.62, 95% CI, 0.46-0.83, p < 0.001), and 68% lower odds of CRC-specific mortality (adjusted OR = 0.32, 95% CI, 0.19-0.51, p < 0.001). Conclusions: Screening colonoscopy is associated with decreased odds of overall and CRC-specific mortality and individuals with older age are more likely to receive it.