Researchers within the RECOVER Initiative share their progress to understand, treat, and prevent Long COVID through research publications. Follow the latest science from RECOVER’s research studies below.
Visit the Research Summaries page to learn about RECOVER’s Long COVID research in a format that’s easy to understand.
Short SummaryTo understand Long COVID, researchers must be able to figure out which patients have it. Our understanding of Long COVID is evolving and it has been difficult to know who had Long COVID, especially in children. We need a reliable method to identify who might have Long COVID using existing health data.
The purpose of this study was to create and test a computer program, called an algorithm, to find out which children have Long COVID based on their electronic health records (EHRs). EHRs (digital medical charts that have health data like doctor visits, lab results, and other health history) are an important source of data for research studies on Long COVID. The algorithm looks at EHRs to find patterns in the diagnoses, prescribed medications, procedures, and lab tests children received after having COVID-19. These patterns can be described as a phenotype, or a set of measured or visible traits, that can tell us who had Long COVID.
The algorithm correctly identified 67% of the patients who had a Long COVID diagnosis from the EHRs. Among the patients who the algorithm said had Long COVID, 91% had a Long COVID diagnosis. Overall, the algorithm was correct in identifying whether a patient had a Long COVID diagnosis 99% of the time. This means the phenotype can be used to recognize which children have Long COVID in EHR data for future studies, or to screen patients to participate in clinical trials.
Short SummaryTo find another way to know if children had COVID, researchers compared health records of 2 groups of children: children who only had a positive antibody test and children who had a positive PCR test. A positive antibody test means a child had COVID in the past, had the COVID vaccine, or both. A positive PCR test means a child currently has COVID. Researchers used data from children’s hospitals in the National Pediatric Learning Health System (PEDSnet) network.
Antibody tests found 2,714 children who had COVID and didn’t have a positive PCR test or COVID vaccine. If researchers used only PCR tests, they wouldn’t have known these children had COVID. Knowing if children had COVID is the first step to know their risk of Long COVID and if symptoms could be related to Long COVID.
Short SummaryRECOVER researchers used data from the RECOVER observational cohort study in adults ages 18 and over. They compared symptoms reported in surveys by participants who did and did not have COVID.
Researchers found 37 symptoms that participants who had COVID reported more often after having COVID compared to participants who never had COVID. 12 of these symptoms could identify participants with Long COVID, including feeling tired and unwell after activity, feeling weak and tired (fatigue), and brain fog. A definition of Long COVID based on symptoms is important for future research, including to find treatments.
Short SummaryA quick way for scientists to identify patterns in a large set of data is by teaching computers to find those patterns for them. To do this, the scientist creates a set of instructions for a computer to follow, called an algorithm, to locate exactly what the scientist is looking for. When that algorithm is plugged into a software program, a computer can run the algorithm many times and learn from it. This process improves the algorithm and it becomes more accurate over time. This is called machine learning. Machine learning can be very helpful and accurate, but it is also challenging to share between researchers because each computer can learn things on its own that can be difficult to recreate. When a computer arrives at an answer that a scientist cannot recreate, it is known as a black-box algorithm.
In order for one research team to share their findings with another research team, they must be able to recreate the steps of the algorithm to “de-black-box" it. This was done by researchers from the National COVID Cohort Collaborative (N3C) as part of the National Institutes of Health (NIH) RECOVER Initiative. The N3C team first identified a phenotype (measured or visible traits) for patients who were at a higher risk of developing Long COVID by creating and training a machine learning-based algorithm. Then, with RECOVER’s support, the N3C researchers worked with researchers from another NIH study called All of Us. They were able to re-create the machine learning-based algorithm, leading to the same phenotype. This means that they were able to “de-black-box" their algorithm. This case can be used as a guide of best practices when sharing algorithm data between research teams. This way, algorithms can be used by many research teams to better understand Long COVID.
Short SummaryAn electronic health record (EHR) is a digital medical chart that has health data like doctor visits, lab results, and other information. These data are useful for understanding trends in health information, including how Long COVID affects people. Because of this, EHR data from different settings can be difficult to compare and use in research.
Researchers from the National COVID Cohort Collaborative (N3C) looked at over 15 million EHRs from 75 hospitals and clinics. Their goal was to make data from different healthcare settings more compatible. To do this, they had to understand and describe how definitions of patient visits differ between healthcare settings so the EHR data from different healthcare settings would be compatible.
The researchers focused on identifying patterns in EHR data. They hoped these patterns would help them gain a better understanding of a patient’s complete care experience, including:
To detect these patterns, researchers created two sets of rules for computer processing of EHR data. These rules are called algorithms. The first algorithm allowed researchers to group EHR data in new ways to make it easier to understand how the information in an EHR is related and easier to analyze individual EHRs and to compare EHRs from different sources.
The second algorithm allowed researchers to identify when EHRs indicated that a patient had been admitted to the hospital. Better data about hospitalizations will help future researchers study COVID and its long-term effects, including Long COVID.
By using algorithms, N3C researchers are trying to make large amounts of EHR data more consistent, manageable, and understandable. Algorithms like the two tested by these researchers can help other researchers enhance the quality of EHR data, making it more consistent and capable of producing important insights about conditions like Long COVID.
