National Clinical Cohort Collaborative (formerly National COVID Cohort Collaborative) (N3C) and NIH RECOVER
Long COVID research using harmonized electronic health records (EHRs) from multiple institutions, including machine-learning phenotyping, ICD-10-based characterization, and target trial emulations of antiviral treatment.
Long COVID can be difficult to identify in health records, especially when infections are not documented or patients have been reinfected. Through N3C and NIH RECOVER, I have worked with collaborators using records from multiple health systems to study Long COVID and Paxlovid treatment for acute COVID-19.
Our work includes updating a machine-learning model to identify patients who may have Long COVID as testing and infection patterns changed. We have also used a target trial framework, which applies the design of a hypothetical trial to observational records, to study Paxlovid. In a study on which I was a co-first author, the estimated effect of Paxlovid on overall Long COVID incidence was not statistically significant; estimates for cognitive and fatigue symptoms suggested small protective effects. The causal interpretation depends on accounting for confounding in the observational data.
Project details
outputs
publications
- Effect of Paxlovid treatment during acute COVID-19 on Long COVID onset: An EHR-based target trial emulation from the N3C and RECOVER consortia
- Re-engineering a machine learning phenotype to adapt to the changing COVID-19 landscape: a machine learning modelling study from the N3C and RECOVER consortia
- Identifying commonalities and differences between EHR representations of PASC and ME/CFS in the RECOVER EHR cohort
- Effect of nirmatrelvir/ritonavir (Paxlovid) on hospitalization among adults with COVID-19: An electronic health record-based target trial emulation from N3C
- Coding Long COVID: Characterizing a New Disease through an ICD-10 Lens
- Identifying Who Has Long COVID in the USA: A Machine Learning Approach Using N3C Data