Objective:
To determine the clinical validity of family history (FH) in colorectal cancer (CRC) risk classification.
Methods:
The Assessment of Risk of Colon Tumors In Canada case-control dataset was used to develop regression models associating risk factors with CRC in Ontario adults. Two regression models (‘clinically-driven’ based on a previously published tool, and data-driven) examined discrimination between CRC cases and controls, with and without the inclusion of FH as a risk variable. Discrimination was assessed using the area under the receiver operator characteristics curve.
Results:
For males, with the addition of FH, there were statistically significant yet quantitatively modest improvements in both models (3.7% clinically-driven, 6.8% data-driven). For females, while FH was a statistically significant predictor of CRC status in the data-driven model, the improvement in discrimination was not significant in either model.
Conclusion:
FH provides very small improvement in model discrimination beyond other standard CRC risk factors.
Identifer | oai:union.ndltd.org:uottawa.ca/oai:ruor.uottawa.ca:10393/31152 |
Date | January 2014 |
Creators | Jonah, Leigh |
Contributors | Wilson, Brenda, Potter, Elizabeth |
Publisher | Université d'Ottawa / University of Ottawa |
Source Sets | Université d’Ottawa |
Language | English |
Detected Language | English |
Type | Thesis |
Page generated in 0.0011 seconds