* Among the 94 patients who did not have a follow-up colonoscopy related to working-up their positive FOBT result, 10 patients ultimately underwent colonoscopy to work-up symptoms that developed over the 7-year period (e.g., hematochezia, unexplained weight loss, anemia) and 2 patients underwent screening colonoscopy many years later without any mention of their positive FOBT result from 2001. The 3 patients in the no follow-up colonoscopy group who died of colorectal cancer all had refused follow-up colonoscopy.⁵
**8 patients did not have a pathology report available from their colonoscopy performed outside of VA to determine size or type of polyps detected. Therefore we were unable to definitively classify them as “significant adenoma” versus “non-significant/normal findings.” None of these 8 patients have any evidence in their medical records that they ever were diagnosed with colorectal cancer. Overall, 17% (35/212) patients received follow-up colonoscopy outside the VA healthcare system.
†Significant adenoma was defined as having an adenoma ≥1 cm, 3 or more adenomas or any adenoma with villous features.
††One patient was diagnosed with an incidental colorectal cancer and survived more than 5 years.
‡12 patients had complications from colonoscopy: three of the 34 patients with significant adenomas had notable bleeding following polypectomy–one required hospital admission; another had a vasovagal episode following epinephrine for bleeding; and one had the procedure aborted and required a repeat colonoscopy. Two patients had complications from other testing after their colonoscopy–one had a fall after a barium enema, requiring transfer to the emergency department but was otherwise unharmed and the other had a sigmoidectomy to treat a large adenoma complicated by a hypoxic event and a 3-week hospitalization. Three of the 45 patients with normal colonoscopies had discomfort with the colonoscopy. Lastly, colonoscopy may have contributed to the death of 1 patient who died within 3 days of his colonoscopy although the exact cause of death is unknown per the medical records. Gray shading indicates patients who potentially benefited from screening.

*Net benefit was assigned to those patients diagnosed with a significant adenoma or colorectal cancer on follow-up colonoscopy who lived at least 5 years after screening, even if they experienced complications from screening or treatment. Net benefit or burden was indeterminate if 1) patients had 1-2 small adenomas and lived at least 5 years or 2) patients had a colonoscopy outside of the VA without an available pathology report and lived at least 5 years . Net burden was assigned to patients who failed to receive colonoscopy after their positive FOBT result, who had a false-positive FOBT result (i.e.,non-significant findings on follow-up colonoscopy) , or who had a significant adenoma detected but died within 5 years . Among patients with best life expectancy who experienced net burden the most common reason was failure to get a follow-up colonoscopy (55%; 18/33) whereas among patients with average or worst life expectancy the most common reason for net burden was dying within 5 years of their FOBT (54%;64/118). 5-year mortality was 5.9% (3/51) for patients with best life expectancy, 38.2% (50/131) for patients with average life expectancy, and 46.7% (14/30) for patients with worst life expectancy which equates to a life expectancy of 5.4 years.