Statistical analysis Comparisons between categorical variables were made using a chi square test and a t-test was used for comparisons between continuous variables. Unconditional MK 1775 logistic regression was used to generate age-, gender-and race-adjusted odds ratios and their 95% confidence intervals (CI) comparing medication users with non-users. Statistical analyses were performed with Inhibitors,research,lifescience,medical SAS 9.3 software. Results Table 1 shows the demographic
characteristics of statin users and non-users. The mean age was higher among statin users than non-users (59.6 vs. 54.4, P<0.0001). A larger proportion of the statin users (58.3%, n=229) vs. non-users (48.1%, n=529) were female (P=0.0005). Users and non-users also differed in their racial breakdown (P=0.01), all metabolic and environmental factors including BMI (P<0.0001),
Inhibitors,research,lifescience,medical presence of diabetes (P<0.0001), hypertension (P<0.0001), smoking status (P=0.03) and alcohol use (P=0.04). Statin users and non-users did not differ in their family Inhibitors,research,lifescience,medical history of colorectal cancer. Table 2 shows the association between statin use and colonoscopy findings, adjusted for age, sex, race, BMI, diabetes, hypertension and smoking or alcohol use. No associations were seen between any of colonoscopy findings and use of statin in the total population or in Hispanics Table 1 Demographics of statin users and non-users Inhibitors,research,lifescience,medical undergoing colonoscopy Table 2 Association between statin use and colonoscopy findings in total population and Hispanics Table 3 shows the demographic characteristics
of aspirin users and non-users. The mean age was higher among aspirin users than non-users (60.0 vs. 54.9, P<0.0001). There was no significant difference between Inhibitors,research,lifescience,medical the groups in terms of gender breakdown. Users and non-users also differed in their racial breakdown (P<0.0001), BMI (P=0.002), presence of diabetes (P<0.0001) and hypertension (P<0.0001). Aspirin users and non-users did not differ in any other factors. Table 4 shows the association second between aspirin use and colonoscopy findings, adjusted for age, sex, race, BMI, diabetes, hypertension and smoking or alcohol use. Compared to non-users of aspirin, those who used aspirin had an increased risk for having two or more adenomas [odds ratio (OR) =1.73, 95% CI: 1.00, 2.99, P=0.05] and presence of an adenoma in the proximal colon (OR =1.66, 95% CI: 1.07, 2.58, P=0.02). No associations were seen between any other colonoscopy findings and aspirin use in the total population. There were no significant associations between aspirin use and colonoscopy findings in Hispanics.