The Inappropriate Use of Fecal Immunochemical Testing in a Safety-Net Hospital

Abstract

Introduction: Denver health and hospital, a large safety-net health care system in Colorado with multiple primary care and subspecialty clinics as well as a 500-bed hospital, switched from guaiac-based stool testing as it’s primary stool-based colon cancer screening modality to fecal immunochemical testing (FIT) in 2012. Studies have demonstrated that FIT is a more sensitive and specific test than guaiac-based stool testing. FIT has only been validated as a screening tool for colorectal cancer in average-risk patients. There is no validation of FIT as a test for gastrointestinal bleeding or in patients who are at high risk for colorectal cancer because of family history. This study examined whether FIT was being used appropriately in the Denver health system. Methods: Denver Health uses the Polymedco OC fit-chek brand of FIT tests. Each analyzed FIT test costs the hospital approximately $14. All of the patients who had a FIT test analyzed in the laboratory of Denver health from April 1, 2012 to March 31, 2013 were identified. Demographic information was collected on the patients. Each FIT test was categorized as either being for screening purposes or diagnostic purposes. Results: In the 1-year study period, 7,986 FIT tests were received in the Denver health laboratory. Of these, 2,442 (30.6%) were ordered for diagnostic purposes (query bleeding), 5,544 (69.4%) were ordered for screening purposes, and 363 (4.5%) were unable to be analyzed for various reasons. The mean age of the patients who had the diagnostic tests was 46±19 years, and the mean age of the patients who had the screening tests was 59±8 years. Nearly all of the analyzed diagnostic tests (2,126, 95%) were ordered in the emergency department, the urgent care clinic, or the inpatient setting. Five hundred fifty-five (25%) of the diagnostic tests were positive. These tests have not led to consistent colonoscopic follow-up. Of the analyzed screening tests, 90 (1.7%) were done inappropriately in the emergency room, urgent care clinic, or inpatient setting. Two hundred sixty-four (5%) of the screening tests were done inappropriately in patients 50 years of age and younger. Conclusion: This is one of few studies looking at the use of FIT in an underserved population. Almost 8,000 FIT tests were returned to the Denver health laboratory over a 1-year period. Of the tests, 4.5% were unable to be analyzed despite being returned to the lab. Nearly 2,500 tests were ordered solely to diagnose bleeding. A small percentage of tests were done in patients less than 50 years of age and in an inappropriate hospital-based setting. Overall, 2,706 tests were done for improper reasons at an estimated cost to the institution of $37,884. This data has led Denver health to remove FIT from the emergency room, urgent care, and inpatient settings.