The lay abstract featured today (for Trivariate Joint Modeling for Family Data with Longitudinal Counts, Recurrent Events and a Terminal Event with Application to Lynch Syndrome by Jingwei Lu, Grace Y. Yi, Denis Rustand, Patrick Parfrey, Laurent Briollais, Yun-hee Choi) is from Statistics in Medicine with the full article now available to read here.
How to cite
Lu, J., Yi, G.Y., Rustand, D., Parfrey, P., Briollais, L. and Choi, Y.-h. (2024), Trivariate Joint Modeling for Family Data with Longitudinal Counts, Recurrent Events and a Terminal Event with Application to Lynch Syndrome. Statistics in Medicine. https://doi.org/10.1002/sim.10210
Lay Abstract
Colorectal cancer (CRC) is the fourth most common cancer and the third leading cause of cancer deaths worldwide. Risks for CRC include poor diet, untreated polyps, aging, and hereditary conditions like Lynch syndrome. Regular screening, particularly colonoscopy with polypectomy to remove polyps, is crucial for early detection and prevention.
In a study of 18 Lynch syndrome families from Newfoundland, most existing models only considered either screening visits or polyp detection, not both. We developed a new model that accounts for both screening and polyp detection simultaneously, while also considering individual and family-level factors.
Because our model is complex, we used the Integrated Nested Laplace Approximation (INLA) method for accurate and efficient calculations. Our results show that this new model provides more precise predictions and highlights the importance of family-level factors.
Our analysis of Lynch syndrome family data showed that more frequent screenings tend to lead to more polyps being found, and finding more abnormal polyps can prompt more frequent screenings. Having more polyps, both individually and within a family, as well as more frequent screenings, were related to a higher risk of CRC. In addition, our model can help identify individuals and families who may need more frequent screenings to better manage their risk of CRC.
This research is valuable because similar studies are rare, and our model handles cases with fewer subjects and incomplete data effectively.
More Details