A research team has developed the Comorbid Operative Risk Evaluation (CORE) score, aiming to better understand how chronic illnesses impact the risk of patients dying after surgery. This score helps surgeons consider existing health issues and assess mortality risk more effectively.
A UCLA research team has developed the Comorbid Operative Risk Evaluation (CORE) score, which aims to enhance the understanding of how chronic illnesses influence the risk of patient mortality following surgery, enabling surgeons to consider existing health problems and more accurately evaluate the risk of death.
For nearly four decades, medical researchers have relied on two primary tools: the Charlson Comorbidity Index (CCI) and the Elixhauser Comorbidity Index (ECI) to assess how pre-existing health issues affect patient outcomes. These tools depend on ICD codes input by healthcare providers and billing staff to reflect patient illnesses. However, they were not specifically designed for surgical patients and tend to focus on chronic conditions that may not be pertinent to those undergoing surgery. Furthermore, they typically gather data from billing records, lacking detailed insights into pre-existing health conditions.
The model was developed using data from 699,155 patients, with 139,831 (20%) designated as the testing group. Researchers examined adults who underwent 62 different operations across 14 specialties from the 2019 National Inpatient Sample (NIS) by using International Classification of Diseases, 10th Revision (ICD-10) codes. They organized ICD-10 codes for chronic illnesses into Clinical Classifications Software Refined (CCSR) categories. Through logistic regression applied to CCSR with significant feature importance across four machine learning techniques, they predicted in-hospital mortality and used the resulting coefficients to determine the CORE score based on a previously validated system. This score ranges from zero, which indicates the lowest risk, to 100, indicating the highest risk.
Research on health services and outcomes utilizing retrospective databases continues to grow in significance within surgical studies. While researchers aiming to expose quality issues and disparities have noble intentions, lacking the right tools may obscure whether adverse results are unrelated to pre-existing health conditions.
“The introduction of innovative statistical software and methods has allowed researchers to leverage extensive databases to address questions related to healthcare quality, disparities, and outcomes,” stated Dr. Nikhil Chervu, a resident physician in the UCLA Department of Surgery and the principal author of the study. “However, these databases mostly extract information from billing records and often miss nuanced data regarding patients’ pre-existing conditions. Without considering variations in chronic illnesses among patients, comparisons across populations may be flawed. Using this score in further research will help validate its effectiveness and enhance the analysis of surgical outcomes using large datasets.”
