Patterns in completion of colorectal cancer screening using fecal immunochemical tests in a health maintenance organization setting

Cameron Haas | 2017

Advisor: Amanda Phipps

Research Area(s): Cancer Epidemiology, Clinical Epidemiology



Fecal immunochemical test (FIT) is a home-based stool test for colorectal cancer (CRC) screening and an effective means to improve rates of CRC screening overall, and thus reduce CRC mortality. We investigated and described patient characteristics associated with completion of FIT once a physician order has been placed.


Enrollees of Kaiser Permanente Washington between ages 50 and 75 years who were given an order for FIT during 2011 through 2012 for CRC screening were observed for receipt of FIT (N=64,148). Patient characteristics were ascertained through administrative and electronic health record data sources. We compared patient characteristics using prevalence ratios, described time from order to return of FIT, and estimated differences in completion of FIT screening with Kaplan-Meier and Cox regression.


Greater than half (53.7%) of all study subjects returned FIT within one year from the date of the first order. The median time to return of FIT for those who completed screening was 13 days from the date of order (mean: 44.5 days: Q1, Q3: 6, 42). Patient factors associated with decreased completion of FIT included: younger age (50-55 years vs. 70-75 years adjusted HR: 0.56; 95% CI: 0.54, 0.59), obesity (vs. normal BMI adjusted HR: 0.82; 95% CI: 0.82, 0.88), and higher Charlson comorbidity index score (3+ vs 0 adjusted HR: 0.87; 95% CI: 0.83, 0.92). There was significant evidence of increased completion of FIT compared to whites among Asian (adjusted HR: 1.36; 95% CI: 1.30, 1.42), Black (adjusted HR: 1.13; 95% CI: 1.07, 1.21), and Hispanic (adjusted HR: 1.12; 95% CI: 1.05, 1.19) race/ethnicities.


We observed greater FIT completion among minority race/ethnicities, suggesting that disparities in CRC screening within these groups is likely due to failure of provider initiation rather than patient completion of the test. However, additional interventions in other groups, such as obese individuals, to improve screening should be considered. Impact: Our results can be used to develop targeted interventions based on patient characteristics that would improve CRC screening compliance.