Predicting Colorectal Cancer and Adenomas Before Colonoscopy
Author Information
Author(s): Adelstein Barbara-Ann, Macaskill Petra, Turner Robin M, Katelaris Peter H, Irwig Les
Primary Institution: University of New South Wales
Hypothesis
Can a model using age and medical history effectively identify individuals at higher risk for colorectal cancer and adenomas?
Conclusion
The study's models can help identify individuals at higher risk for colorectal cancer and adenomas, allowing for prioritized colonoscopy.
Supporting Evidence
- The model achieved an area under the ROC curve of 0.83 for cancer prediction.
- Age was identified as the dominant risk factor for both cancer and adenomas.
- Having a colonoscopy in the last 10 years significantly reduced the risk of cancer and advanced adenomas.
- The study included over 8,000 patients, providing a robust dataset for analysis.
- Symptoms like rectal bleeding and mucus were associated with higher cancer risk.
Takeaway
This study helps doctors figure out who is more likely to have serious problems in their bowels, so they can get checked sooner.
Methodology
Patients completed a questionnaire about their symptoms and medical history before undergoing colonoscopy, and multinomial logistic regression was used to analyze the data.
Potential Biases
The study may not be applicable to patients presenting in general practice, as it focused on a referred population.
Limitations
There was no standard reporting for colonoscopy, which may affect the consistency of findings.
Participant Demographics
47% male, age range 18 to 95 years, with a median age of 58 years.
Statistical Information
P-Value
0.51
Confidence Interval
95%
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website