BACKGROUNDStudies exploring CRC testing prevalence and correlates within US populations have provided limited and sometimes conflicting information. The most recent national-level reports have described US usage of CRC tests but none have considered only those tests done specifically for screening reasons as an outcome variable.METHODSUsing the NHIS 2000 sample of >or=50 year-old, we assessed screening behavior using an outcome variable accounting for (1) any combination of recommended tests (2) done within their respective time guidelines, and (3) specifically for screening purposes.RESULTSOnly 25.8% (95% CI: 24.9-26.7%) of the population reported getting a test done for screening purposes within the recommended time. Most (>85%) of the FOBTs and only about 60% of endoscopies were done for screening. Among those who had an endoscopy within the recommended time, Blacks were more likely than Whites to report screening as the purpose of the test. Hispanics had the lowest test usage irrespective of test time, reason or type. Hispanics were 50% (p<0.001) less likely to be adherent, and Blacks approximately 22% (p<0.01) less likely to be adherent, than Whites. After multivariate adjustment, differences between Whites and Blacks disappeared; Hispanics remained less likely. Increasing education predicted higher adherence among Whites but only undergraduate completion did so among Blacks. Male gender predicted adherence only among Blacks and insurance only among Hispanics.CONCLUSIONSPreventive screening for CRC is lower than estimates from previous studies. Future studies should consider accounting for test purpose. Our findings need confirmation through studies based on objective data.