Design of the Cancer Care Outcomes and Surveillance Consortium (CanCORS): Methodologic Challenges


The CanCORS project enrolled and followed a cohort of approximately 10,000 patients with lung or colorectal cancer between September 2011 and January 2014. The goal of this observational study was to learn, in a “population-based” cohort, about access to and patterns of care, patient and physician perceptions of care, and outcomes. The Coordinating Center faced several challenges in the design, including attempting to construct a representative sample from a sampling frame dictated by the NCI, setting uniform data collection standards in many data collection sites and call centers, providing a uniform and principled way to analyze survey data subject to item response missingness, and preserving confidentialitywhen releasing data where the risks of re-identification could be high for some subjects. In this talk, we will discuss the use of central multiple imputation in data that will be used for many purposes and by many analysts, and the use of syntheticdata in potential public use data sets.

Additional Information

View slides from this presentation.

Date & Time(s)


Memorial Sloan Kettering Cancer Center
1275 York Avenue
Room M-107
New York, NY 10065


Survivorship, Outcomes, And Risk Program


David P. Harrington, PhD
Professor of Biostatistics
Department of Biostatistics, Harvard School of Public Health
Department of Biostatistics and Computational Biology, Dana Farber Cancer Institute
Boston, MA