Challenges and Approaches to Measuring Repeat Fecal Immunochemical Test for Colorectal Cancer Screening

Research output: Contribution to journalArticle


BACKGROUND: Colorectal cancer screening with fecal immunochemical testing (FIT) can reduce colorectal cancer-related mortality. Effectiveness of FIT may be compromised when patients do not adhere to a regular schedule. However, having no standard measure of repeat FIT presents challenges for assessing effectiveness across populations and settings. We compared three measures of repeat FIT in a large, integrated health care system in Dallas, Texas. METHODS: We identified 18,257 patients age-eligible (50-60 years) for FIT in January 1-December 31, 2010 and followed over four rounds of screening. Measures included: (i) repeat FIT in prior screeners, or completion of FIT within 9-15 months of the previous; (ii) yes-no patterns, whereby patients were assigned yes or no in 9-15 month windows; and 3) proportion of time covered (PTC), or the amount of time patients were up-to-date with screening relative to time eligible. RESULTS: Repeat FIT varied by measure. Using a prior screeners measure, 15.8% of patients with a normal FIT in round 1 completed repeat FIT in round 2. Repeat FIT was notably higher (52.3%) using PTC. The most common yes-no pattern was YNNN or "one-and-done," and only 9.4% of patients completed two consecutive FITs across all rounds (YYNN). CONCLUSIONS: Different measures of repeat FIT yielded a range of estimates, making comparison across studies difficult. Researchers should weigh the advantages and disadvantages of each measure and select the most appropriate to their research question. IMPACT: Our study highlights the need for future research of repeat FIT measures that best approximate screening effectiveness.

ASJC Scopus subject areas

  • Epidemiology
  • Oncology

Fingerprint Dive into the research topics of 'Challenges and Approaches to Measuring Repeat Fecal Immunochemical Test for Colorectal Cancer Screening'. Together they form a unique fingerprint.

  • Cite this