We used data from the 2008 Oregon Behavioral Risk Factor Surveillance System (BRFSS), a statewide, random-digit-dialed telephone survey of health conditions and risk behaviors of the noninstitutionalized Oregon population aged 18 years or older, to estimate family history of CRC, health care provider practice, CRC screening, and preventive behavior. Detailed information about the Oregon BRFSS is available elsewhere (13
). We weighted the data by age and sex to better reflect the demographic characteristics of adults in Oregon. The Oregon Health Authority deemed projects that use BRFSS data, including this project, to be exempt from review.
For this study, we classified respondents as having a family history if they reported at least 1 first-degree relative (parent, sibling, or child) with CRC and as having a family history of early-onset CRC if they reported at least 1 first-degree relative who was diagnosed with CRC before the age of 50. We classified respondents who reported no first-degree relatives with CRC or who reported they were adopted and did not know the medical history of their biological family members as having no family history.
Additionally, we collected information about respondent-reported clinician recommendations on CRC risk and screening, as well as discussions about prevention and behaviors that might affect CRC risk. We also collected information about whether or not patients acted on clinicians' recommendations.
We asked respondents, "Have you ever been told by a doctor, nurse, or other health care provider that you have colorectal cancer?" Respondents without CRC were asked the following questions:
"Thinking of your close blood relatives, do you have a parent, brother or sister, or child who has been diagnosed with colorectal cancer by a health care provider?" If they responded yes, they were asked to identify which of these relatives had been diagnosed with CRC and to report the number of relatives who had been diagnosed with CRC before the age of 50.
"Has a doctor, nurse, or other health care provider ever discussed the chances of you getting colorectal cancer?"
"Has a health care provider ever discussed testing for colorectal cancer with you?"
"Has a health care provider ever recommended changes in eating habits or physical activity to reduce your chances of getting diseases like colorectal cancer?"
"How likely do you think it is that you will get colorectal cancer in the future?"
"Have you made changes in your eating habits or physical activity to reduce your chances of getting diseases like colorectal cancer?"
Respondents without CRC and who were aged 50 or older were asked whether they had ever had a CRC screening test, including fecal occult blood test (FOBT)Respondents without CRC and who were aged 50 or older were asked whether they had ever had a CRC screening test, including fecal occult blood test (FOBT), sigmoidoscopy, or colonoscopy and the length of time since each test. We analyzed CRC screening prevalence by using 2 definitions: 1) ever having a colonoscopy, and 2) having CRC screening within the recommended time period based on the American Cancer Society and US Multi-Society Task Force on CRC guidelines for average-risk populations, which is either an FOBT within the past year, sigmoidoscopy within the past 5 years, or colonoscopy within the past 10 years (10
We analyzed the following potential covariates, which may affect associations among family history, health care provider recommendations, and patient behavior. Covariates included self-reported information on age, sex, education level, annual household income, marital status, race/ethnicity, leisure-time physical activity within the past month, obesity, smoking status, alcohol use, insurance status, and having 1 person the respondent thought of as their personal doctor or health care provider. We defined obesity as having a body mass index of 30 kg/m2 or more, current smokers as people who reported smoking every day or some days and who reported having smoked 100 or more cigarettes during their lifetime, heavy alcohol use as consuming more than 2 alcoholic drinks per day on average for men and more than 1 alcoholic drink per day on average for women within the past month.
We used Pearson χ2 tests and logistic regression to assess the association between respondents' family history status and reported health care provider practices, perceived risk for developing CRC, preventive and screening behaviors, and risk factors for CRC.
We included only covariates that were significantly associated with family history and the outcome variable in bivariate analyses in the multivariable logistic regression models. In the adjusted logistic regression models, we kept only covariates that changed the point estimate of the odds ratio (OR) by at least 10% (compared with the full model) in the final models. All analyses were performed using Stata version 11.0 (StataCorp LP, College Station, Texas). We reported sample sizes (number of survey respondents) as unweighted numbers and percentages as weighted estimates.
Of the 1,841 people who responded to the family history questions, we excluded 29 because of missing or unknown information about family history of CRC and 17 additional respondents who had CRC. Our final sample for this analysis included 1,795 respondents without CRC. Although we were able to present the overall prevalence of respondents with 2 or more relatives with CRC, we were unable to stratify this group by other variables because of the small number of respondents (n = 9).