Date de publication
Nom du journal
Cancer epidemiology
Background: The quantitative faecal immunochemical test (FIT) detects pre-cancerous lesions and colorectal cancer (CRC). Current CRC screening programmes are based upon a binary FIT result. This study evaluates the possibility of personalised risk prediction, based on FIT, age and gender to gain more insight into improving discrimination between normal outcomes, pre-cancerous lesions, carcinoma in situ and CRC working towards a more tailored screening approach.
Methods: In this retrospective study, from October 2013 until July 2016, data from 57,421 participants who underwent a colonoscopy after a positive FIT in the Flemish CRC screening programme were analysed using a multinomial multivariable logistic regression model.
Results: A significant difference in risk of detecting neoplasia was found between the established risk profiles based on the combination of the quantitative FIT, age and gender. The odds for detecting CRC in men aged 74, with a FIT result of ≥1000 ng/ml, was higher by a factor of 58.43 than that for women aged 56, with a FIT result of 75 ng/ml.
Conclusion: A large difference in risk with regard to the detection of colorectal neoplasia was found, based on demographics of the population. For some participants, the chance of finding no anomalies was more than 60%. Including additional variables in a prediction model could further increase discrimination between outcomes and practicality.
Methods: In this retrospective study, from October 2013 until July 2016, data from 57,421 participants who underwent a colonoscopy after a positive FIT in the Flemish CRC screening programme were analysed using a multinomial multivariable logistic regression model.
Results: A significant difference in risk of detecting neoplasia was found between the established risk profiles based on the combination of the quantitative FIT, age and gender. The odds for detecting CRC in men aged 74, with a FIT result of ≥1000 ng/ml, was higher by a factor of 58.43 than that for women aged 56, with a FIT result of 75 ng/ml.
Conclusion: A large difference in risk with regard to the detection of colorectal neoplasia was found, based on demographics of the population. For some participants, the chance of finding no anomalies was more than 60%. Including additional variables in a prediction model could further increase discrimination between outcomes and practicality.