The RECOVER initiative pools real world data (RWD) from participating PCORnet sites and utilizes advanced machine learning and artificial intelligence methods to better understand the long-term health effects of COVID-19 after the acute infection period. This is achieved through the development of condition specific computable phenotypes which will have the capacity to detect patients that are potentially affected by PASC and potentially identify patients that are susceptible to PASC. This project also focuses on health services research (HSR) which investigates other confounding factors that can play a role in clinical outcomes and treatment such as racial/ethnic disparities, environmental factors, and analyzing how different variants of COVID-19 have affected individuals. This projects overarching goals are described below:
- What are the clinical spectrum of and biology underlying recovery from acute SARS-CoV-2 infection over time?
- For those patients who do not fully recover, what is the incidence/prevalence, natural history, clinical spectrum, and underlying biology of this condition? Are there distinct phenotypes of patients who have prolonged symptoms or other sequalae?
- Does SARS-CoV-2 infection initiate or promote the pathogenesis of conditions or findings that evolve over time to cause organ dysfunction or increase the risk of developing other disorders?
For more information on this initiative please visit: https://recovercovid.org