Work

Patient and physician preferences among colorectal cancer screening tests: updated predictions from a discrete choice experiment

Year: 2025

Type: article

Abstract: Objective Colorectal cancer (CRC) screening rates fall short of national goals. This study aimed to understand eligible individuals’ preferences regarding CRC screening modalities to improve screening adherence.Methods Two cohorts took a discrete choice experiment survey: adults aged 45–75 years at average risk for CRC, and physicians who recommend CRC screening. Five attributes (test type, test frequency, true positive rate, true negative rate, adenoma true positive rate) from four different CRC screening modalities (cell-free DNA blood test [cf-DNA-BT], colonoscopy, fecal immunochemistry test [FIT], multi-target stool DNA [mt-sDNA] test) were assessed. Test-specific performance was derived from clinical data. Predicted choice probability (PrCP) of each modality was calculated from the results of a mixed logit model. Subgroup analyses were performed.Results Among 1,249 adult respondents, mt-sDNA was the preferred modality (PrCP 39.4%) versus colonoscopy (24.8%), cf-DNA-BT (21.1%), and FIT (14.7%). In all subgroups, respondents preferred mt-sDNA. Respondents with previous noninvasive CRC screening experience (PrCP 53.5%) and with no prior CRC screening (42.1%) preferred mt-sDNA. PrCP for colonoscopy was lower for populations historically less likely to be adequately screened for CRC (such as 45–49-year-old respondents [PrCP 21.1%] and non-White respondents [23.1%]). PrCP among 400 physicians was highest for colonoscopy (PrCP 95.0%) versus mt-sDNA (4.3%), cf-DNA-BT (0.4%), and FIT (0.3%). No significant difference in preferences was found between primary care providers and gastroenterologists.Conclusion US adults eligible for CRC screening preferred mt-sDNA testing over other screening modalities, while physicians highly preferred colonoscopy. Offering CRC screening options that align with patient preferences may result in higher screening adherence.

Source: Current Medical Research and Opinion

DOI: 10.1080/03007995.2025.2576596