1039 — Testing and System Interventions Improve Colorectal Cancer Screening Uptake
Hoffman RM (Albuquerque VA Medical Center), Yee EF
(Albuquerque VA Medical Center), Massie L
(Albuquerque VA Medical Center), Steel S
(Albuquerque VA Medical Center), Schrader RM
(University of New Mexico School of Medicine), Murata GH
(Albuquerque VA Medical Center)
Compare adherence with fecal immunochemical testing (FIT) vs. guaiac-fecal occult blood testing (gFOBT) for colorectal cancer (CRC) screening. Determine whether a population-based strategy of mailing stool test cards to patients due for screening might achieve higher screening uptake than visit-based screening.
We used the electronic medical record (EMR) to identify 7,053 New Mexico Veterans Affairs patients due for CRC screening in 2008, and randomly selected 3,869 to receive mailed invitations to participate in a trial comparing FIT and gFOBT. We asked interested subjects to return a postcard and called respondents to confirm eligibility. We randomly allocated 404 subjects to receive either FIT or gFOBT by mail. In identifying these subjects, we created 3 control groups of patients who would require visit-based screening--the initial 3,184 randomly-assigned controls (CG1), 2525 patients who did not respond to invitations to participate in the randomized trial (CG2), and 257 respondents who expressed interest in participating but could not be contacted (CG3). For the randomized trial, we used Chi-squared tests to compare adherence for completing fecal blood testing within 3 months. For the comparative effectiveness analysis, we measured gFOBT screening within 3 months after enrollment in the intervention group, and gFOBT or colonoscopy screening within 6 months of being identified as a control subject. We compared screening uptake across groups using multivariate logistic regression to adjust for race/ethnicity, gender, clinic site, previous gFOBT testing, and comorbidities.
Screening adherence was higher with FIT than gFOBT (61.4% vs. 50.5%, P = 0.03). Subjects mailed a gFOBT (48.5%) achieved higher CRC screening uptake than control groups (18.5% for CG1; 14.3% for CG2; 18.8% for CG3). Adjusted odds ratios for screening uptake among the control groups were all less than for the intervention group (OR = 0.22, 0.17, 0.16, respectively, P-values < 0.0001).
Testing adherence was higher with FIT than gFOBT. Using an EMR to initiate CRC screening appeared more effective than relying on clinic visits to achieve screening.
Our VA subsequently adopted the FIT for CRC screening. Using an EMR to implement a non visit-based, population approach to cancer screening supports the Veteran-Centered Medical Home model.