The primary objective of this study was to determine the prevalence of colorectal cancer (CRC) screening among eligible patients in a large medical practice. A secondary objective was to compare CRC screening rates obtained from medical records with physician self-reported CRC screening recommendation. We conducted a retrospective record review of 214 patients ages > or = 50 years of a large multispecialty medical organization in Houston, Texas, for receipt of fecal occult blood test (FOBT), flexible sigmoidoscopy (SIG), and/or colonoscopy (COL). We estimated prevalence using two definitions: (a) FOBT in past year or SIG in past 5 years or COL in past 10 years; and (b) FOBT in past year and SIG in past 5 years or COL in past 10 years. Age, gender, race/ethnicity, family history, number of chronic conditions, and index visit were independent variables. Contingency table and logistic regression analysis were used to test for associations between outcomes and independent variables. Our study population was 48% male with a mean age of 63 years (range: 53-84 years). One-quarter of the records showed FOBT by 3-day kit (51 of 214) and 27% by digital rectal exam (57 of 214). SIG was recorded in 32% of records. Half (54%) of the records had documentation of CRC screening according to definition no. 1 and 19% according to definition no. 2. Screening rates from medical record review were lower than those derived from physician self-report. Our findings underscore the need for interventions to improve CRC screening in primary care settings.