Evaluation of mCODE Coverage in EHR: a Scoping Review of Cancer Natural Language Processing

Liwei Wang, Sunyang Fu, Andrew Wen, Xiaoyang Ruan, Huan He, Sijia Liu, Sungrim Moon, Michelle Mai, Irbaz Bin Riaz, Nan Wang, Ping Yang, Hua Xu, Jeremy L. Warner, Hongfang Liu

Abstract: Studies have shown structured data in Electronic health records (EHRs) alone are not sufficient for cancer research and clinical care. In this review, we aim to assess the coverage of Minimal Common Oncology Data Elements (mCODE) in EHR by elucidating the alignment of NLP extracted data elements with this oncology data standard. We conducted a scoping review based on methods proposed by Arksey and O'Malley. In total 123 publications were selected and included in our analysis. Results show that that some critical data elements covered by mCODE are not fully captured as part of structured EHR data. In addition, cancer research and patient care require some data elements beyond mCODE.