Fig. 1

Using NLP and AI to guide VBC. Suggested methodology to analyze clinical outcomes in VBC using NLP. Data from single or multiple hospital sites are extracted and analyzed from anonymized EHRs using NLP. The structured and unstructured data are aggregated into a single database, where the target clinical population is screened. Patients using a specific drug (Drug X) can be extracted from all patients eligible for a specific pharmacological agent. Several clinical outcomes and metrics (as specified in the VBC) can be evaluated at the population level in those patients prescribed with the drug. This real-world information can be then used to adjust and negotiate pricing. EHR = electronic health record; LDL-C; low-density lipoprotein cholesterol; MI = myocardial infarction; ML = machine learning; NLP = Natural Language Processing; VBC = value-based contract