Development of a natural language processing algorithm to extract social determinants of health from clinician notes.

Abstract

Disparities in access to the organ transplant waitlist are well-documented, but research into modifiable factors has been limited due to lack of access to organized pre-waitlisting data. This study aimed to develop a natural language processing algorithm to extract social determinants of health from free text notes and quantify the association of SDOH with access to the transplant waitlist. We collected 261,802 clinician notes from 11,111 adults referred for kidney or liver transplant between 2016-2022 at Duke University Health System. A social determinants of health ontology and a rule-based natural language processing algorithm were created to extract and organize terms. Education, transportation, and age were the most frequent terms identified. Negative sentiment and refer were the most negatively associated features with listing in both kidney and liver transplant patients. Income and employment for kidney, and judgment and positive sentiment for liver were the most positively associated features with listing. This study suggests that the integration of natural language processing tools into the transplant clinical workflow could help improve collection and organization of social determinants of health and inform center-level efforts at resource allocation, potentially improving access to the transplant waitlist and post-transplant outcomes.