Abstract
INTRODUCTION: Emergency department (ED) clinicians utilize the guaiac fecal occult blood test (gFOBT) in their assessment of suspected gastrointestinal bleeding or unexplained anemia despite supporting evidence. ED clinicians' ability to predict the gFOBT results and how the gFOBT results could affect ED patient disposition has not been previously studied. Methods: From October 16, 2019, through September 15, 2020, we conducted a single-site survey of ED clinicians before and after performing gFOBTs during routine clinical care. Survey data were collected and retrospectively evaluated with unadjusted and multivariable regression analyses. RESULTS: We examined a total of 133 combined pre-gFOBT and post-gFOBT surveys. ED clinicians accurately predicted gFOBT results with an area under the receiver operating characteristic curve of 0.75 (95%CI, 0.66-0.85). Of clinician-predicted certain or very probable positive gFOBT results, only 79% were actually positive. In multivariable analyses, decreased hematocrit level (odds ratio (OR), 0.31/10% increase; 95%CI, 0.14-0.61), decreased red blood cell count (OR, 0.41/1x10(12)/L increase; 95%CI, 0.21-0.75), and absence of firm stool consistency (OR, 0.09; 95%CI, 0.01-0.42) were associated with positive gFOBT results (all P<.006). The most common reason for performing gFOBTs was black stool or suspected melena, followed by decreased hemoglobin level, red blood in stool, and suspected upper gastrointestinal tract bleeding. Before performing gFOBT, 50.8% of clinicians responded that the test results would change patient disposition, which decreased to 30.5% after the gFOBT result. CONCLUSIONS: We found that ED clinicians cannot predict the gFOBT results with high accuracy. A suspected GI bleed is the main reason for performing the test in the ED.