Demographic, medical, and behavioral factors associated with colorectal cancer screening and diagnostic colonoscopy in Israel.

557 Background: Colorectal cancer (CRC) is the third leading cause of cancer death worldwide and second leading cause of cancer death in Israel. Israel's universal health care system funds CRC screening (CRCS) (fecal occult blood testing or colonoscopy) for all adults aged 50-74. CRCS adherence is influenced by numerous factors including patient characteristics. In addition to screening asymptomatic individuals, those with anemia may have undiagnosed CRC and require aggressive case finding. This study’s goals were: 1) To compare the socio-demographic, health status, and health behavior characteristics of the eligible Israeli population who underwent any versus no CRCS (2009-2012); 2) To examine the relationship between anemia severity and CRCS likelihood; 3) To determine whether anemia status predicts time-to-CRCS. Methods: National retrospective cohort study of Israeli adults aged 51-71 on 1.1.2009 using electronic medical records from all 4 Israeli HMOs (n=1,009,898) over a 4-year period. Results: Differences in CRCS adherence were observed by patient characteristics. For example, ever-smokers comprised 17.9% of the screened population and 24.9% of the non-screened population (p<0.001). Increasing anemia severity was positively associated with any CRCS performance, especially colonoscopy. For example, severely anemic individuals (hemoglobin <10 g/dL for men, <8 g/dL for women) were significantly more likely to receive colonoscopy than those who were not anemic (OR = 3.23; 95% CI, 2.58-4.27). Anemia somewhat expedites CRCS. For example, 8.9% of those with anemia and 7.6% of those without anemia performed any CRCS within 30 days of hemoglobin level recording (p<0.001). Conclusions: Full CRCS adherence is not yet optimized in Israel. Particular attention directed at those at-risk for CRCS non-adherence may decrease CRC mortality. Although lower hemoglobin somewhat expedites colonoscopy performance, few individuals with anemia actually undergo colonoscopy. Further analysis accounting for confounding and effect modification is crucial for a more complete understanding of these relationships.