Monday 16 May 2022
Using machine learning, researchers are finding patterns in electronic health records to better identify people who may have the disease.
A research team supported by the National Institutes of Health has identified the characteristics of people with long-term COVID and those who are likely to have had it. Scientists have analyzed the unprecedented array of electronic health records (EHRs) available for COVID-19 research using machine learning techniques to better determine who has long-held COVID. To examine EHR data not specified in the National COVID Cooperative Group (N3C), a central national public database maintained by the National Center for the Advancement of Translational Sciences (NCATS) of the National Institutes of Health, it used team data to more than 100,000 likely long-term COVID cases in October 2021 (as of May 2022, the number was over 200,000). The results appear in Digital Health The Lancet†
Long-term COVID-19 is characterized by widespread symptoms, including shortness of breath, fatigue, fever, headache, “brain fog” and other neurological problems. These symptoms can persist for several months or more after the initial diagnosis of COVID-19. One of the reasons COVID has been so difficult to identify for so long is that many of its symptoms are similar to those of other diseases and conditions. Better characterization of COVID-19 could lead to a better prognosis and new therapeutic approaches.
“It makes sense to leverage modern data analytics tools and a unique big data source like N3C, where many of COVID’s long-standing features can be represented,” said study co-author Emily Pfaff, PhD, clinical information scientist. at University. From North Carolina to Chapel Hill.
The N3C data pocket currently contains information from more than 13 million people across the country, including nearly 5 million positive cases of COVID-19. The resource enables rapid research into new questions related to COVID-19 vaccines, treatments, risk factors and health outcomes.
The new research is part of a larger National Institutes of Health initiative, COVID Research to Support Recovery (Recovery), which aims to improve understanding of the long-term effects of COVID-19, called post-acute consequences of SARS. . CoV-2 (PASC) infection. RECOVER will accurately identify people with PASC and develop prevention and treatment approaches. The program also answers key research questions about the long-term effects of COVID through clinical trials, longitudinal observational studies, and more.
in the Lancet Study, Pfaff, Melissa Heindel, PhD, of the University of Colorado Anschutz Medical Campus, and colleagues looked at patient demographics, health care utilization, diagnoses, and medications in health records of 97,995 adult patients with COVID-19 in N3C. They used this information, along with data from nearly 600 long-term COVID patients from three long-term COVID clinics, to create three machine learning models to identify long-term COVID patients.
In machine learning, scientists “train” computational methods to quickly sift through large amounts of data to reveal new insights — in this case about long-term COVID. The models looked for patterns in the data that could help researchers understand patient characteristics and better identify people with the disease.
The models aimed to identify potential long-term COVID-19 patients among three groups in the N3C database: all COVID-19 patients, patients hospitalized with COVID-19, and patients who had COVID-19 but were not in hospital. were hospitalized. The models proved to be accurate, as people at long-term risk for COVID were comparable to patients seen in long-term COVID clinics. The machine learning systems classified nearly 100,000 patients in the N3C database whose profiles were identical to those with a long-term outbreak of COVID-19.
Josh Wessel, MD, PhD, senior clinical advisor at NCATS and a leading program scientist in Recover, said. “Was there anything different about these people long before they had COVID? Do they have certain risk factors? Was there anything about the way they were treated during acute COVID that could increase or decrease the risk of long-term COVID infection? †
The models looked for commonalities, including new medications, doctor visits and new symptoms, in patients with a positive COVID diagnosis who were at least 90 days post-acute infection. Models identified patients with long-term COVID if they attended a long-term COVID clinic or had long-term COVID symptoms and likely had the disease but were undiagnosed.
“We want to take the new patterns we see with the COVID diagnostic code and incorporate them into our models to try to improve their performance,” said Handel of the University of Colorado. “Models can learn from a larger group of patients and become more accurate. We hope we can use our long-term COVID patient classification to recruit participants in clinical trials. †
This study was funded by NCATS, which contributed to the design, maintenance, and security of the N3C enclave, and the NIH RECOVER initiative, with support from NIH OT2HL161847. Recover coordinates, among other things, the participant recruitment protocol to which these activities contribute. Analyzes were performed with data and tools accessible through NCATS N3C Data Area. Supported by NCATS U24TR002306.
About the National Center for the Advancement of Translational Sciences (NCATS): NCATS conducts and supports research in translational science and function – the process by which health-enhancing interventions are developed and implemented – to enable more treatments to reach more patients faster. To learn more about how NCATS can help shorten the journey from scientific observation to clinical intervention, visit https://ncats.nih.gov.
About the National Institutes of Health (NIH):NIH, the nation’s medical research agency, is made up of 27 institutes and centers and is part of the United States Department of Health and Human Services. The National Institutes of Health is the premier federal agency that conducts and supports basic, clinical, and polymedical research, exploring the causes, treatments, and cures for common and rare diseases. For more information about the National Institutes of Health and its programs, visit www.nih.gov.
National Institutes of Health… Transforming Discovery into Health