Abstract
Abstract Context: The use of predictive analytics can help health systems target the right services to the right patients at the right time while improving population health. Multilevel data, or data at the interpersonal, organizational, community, and policy levels, is rarely sought after but may be used to improve risk prediction by providing information about a patient and the many groups to which they belong. Colorectal cancer screening promotion can be expensive and not all patients need it. Objective: This study sought to develop a risk prediction model using individual data available in the EHR, and then compared a multilevel risk prediction model using publicly available multilevel data. Study design and analysis: Retrospective data was used to develop a risk model, using logistic regression, first in data available in the EHR, and then adding multilevel data. Setting or dataset: This retrospective data only study was conducted at Kaiser Permanente Northwest. Population studied: Patients were eligible who were 50-75 years old, had been members at KPNW for at least a year before their birthday in 2018, and if they were due for screening. Results: The individual level model was sufficient and included 14 characteristics, and had a bootstrap corrected C-statistic of 0.722. The multilevel model added 9 variables, and the bootstrap corrected C-statistic was 0.724. Conclusions: Models can be used to identify patients unlikely to screen, although the performance of the model after adding multilevel data did not improve. Screening models could help in population health management of CRC screening. Citation Format: Amanda F. Petrik. The development of a multilevel risk prediction model to identify risk of failure to complete CRC screening [abstract]. In: Proceedings of the AACR Special Conference on Colorectal Cancer; 2022 Oct 1-4; Portland, OR. Philadelphia (PA): AACR; Cancer Res 2022;82(23 Suppl_1):Abstract nr A001.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
