1046 — The Contribution of Patient Background, Cognitive, and Environmental Factors to Colorectal Cancer Screening Adherence
Partin MR (Center for Chronic Disease Outcomes Research, Minneapolis VAMC and University of Minnesota), Noorbaloochi S
(Center for Chronic Disease Outcomes Research, Minneapolis VAMC and University of Minnesota), Grill J
(Center for Chronic Disease Outcomes Research, Minneapolis VAMC), Burgess DJ
(Center for Chronic Disease Outcomes Research, Minneapolis VAMC and University of Minnesota), van Ryn M
(University of Minnesota), Fisher DA
(Durham Veterans Affairs Medical Center, and Duke University Medical Center), Griffin JM
(Center for Chronic Disease Outcomes Research, Minneapolis VAMC and University of Minnesota), Powell AA
(Center for Chronic Disease Outcomes Research, Minneapolis VAMC and University of Minnesota), Yano E
(VA Greater Los Angeles HSR&D Center of Excellence for the Study of Healthcare Provider Behavior ), Vernon SW
(University of Texas-Houston)
We assessed the relative contribution of background (demographic, health characteristics), cognitive (knowledge, attitudes), and environmental factors (social and medical support for screening) to colorectal cancer (CRC) screening adherence.
The sample included 3,025 patients from 24 VA facilities responding to a national survey of veterans age 50-75 (response rate 83%) administered by mail (with phone follow-up). Survey data on screening behaviors, attitudes, and demographics were linked to: organizational data on screening promotion strategies from the 2007 Clinical Practices Organizational Survey; facility complexity scores; and diagnoses and screening history from VA administrative data. Hierarchical logistic regressions examined the association between adherence (fecal occult blood test in the past 15 months, sigmoidoscopy or double contrast barium enema in the past 5.5 years, or colonoscopy in the past 11 years) and background, cognitive, and environmental factors.
Adherence rates were 83% overall (75% among individuals with no bowel disease history), but varied considerably across facilities (66-96%). The effects of background factors on adherence were mediated by cognitive and environmental factors, and cognitive and environmental factors contributed equally to adherence rates. Perceived importance of CRC screening (OR 1.76, p = .006) and self-efficacy (OR 1.35, p = .01) were significant independent predictors of adherence, but few patients lacked these facilitators. Physician recommendation (a measure of medical support) was the strongest independent predictor of adherence (OR 2.82, p < .0001), but only 16% of patients did not receive a recommendation. Other medical support factors independently associated with adherence included use of clinical champions (OR 1.71, p < .0001) and primary care provider training (OR 1.28, p = .0007) to promote CRC adherence, and organizational complexity (OR 1.46 for lowest vs highest complexity, p = .008). Facilities with the lowest adherence rates had significantly fewer patients receiving a physician recommendation (80 vs 86%, p = .03), and lower rates of colonoscopy use (34 vs 56%, p = .005).
Targeting facilities with low adherence rates may be a more efficient approach to increasing CRC screening adherence in the VA than implementing broad interventions to address patient barriers.
Increasing colonoscopy capacity and addressing provider barriers to recommending screening could be fruitful intervention targets in facilities with low adherence rates.