Evidence map›Paper›PMID 40858480›Full record

ArticleThe Lancet. Digital health2025

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.

Miles Crosskey, Tomas McIntee, Sandy Preiss, Daniel Brannock, John M Baratta, Yun Jae Yoo, Emily Hadley, Frank Blanceró, Robert Chew, Johanna Loomba and 6 more

Abstract read
PubMed Full text at PMC Publisher
In one paragraph

Article in The Lancet. Digital health, 2025. The graph could read no effect estimate from its abstract, so it casts no vote on the map. Cited by 8 papers.

0numbers the graph read from it
0cells of the map it votes in
8citing papers in PubMed
–field-weighted citation impact
1 · What the graph read from it

What it found

Each row is one number read from the abstract, on the scale the paper reported it, with its interval. Left of the dashed line favours the treatment, right favours the comparator. Under each row is the sentence it came from. New to these charts? A ten-minute tutorial.

The abstract states no effect estimate the extractor could read, or names no intervention and outcome on the map, so this paper lights no cell and moves no belief. It is still indexed, cited and linked below.

2 · The registry

The trial behind it

Trials whose registry record cites this paper, or whose number appears in the abstract. A trial that started after this paper was published is citing it as background, not reporting it.

Neither the registry nor the abstract names a trial number. If this is a trial report, that itself is worth knowing.

3 · Its place in the literature

Who cites it

8 citing papers in PubMed.

  1. Article
  2. Metformin and Severe Post-COVID-19 Outcomes Among Individuals with Diabetes Mellitus.medRxiv : the preprint server for health sciences · 2026
    Article
  3. Article
  4. Long COVID Incidence Proportion in Adults and Children Between 2020 and 2024: An Electronic Health Record-Based Study From the RECOVER Initiative.Clinical infectious diseases : an official publication of the Infectious Diseases Society of America · 2025
    Article
  5. Article
  6. Article
  7. Article
  8. Review
4 · The record

Corrections and comments

PubMed lists nothing against this paper. Absence here is not a guarantee, only a check that was made.

5 · Who and what money

Authors and funding

16 authors.

Miles CrosskeyCoVar Applied Technologies, Durham, NC, USA.
Tomas McInteeNC TraCS Institute, University of North Carolina at Chapel Hill School of Medicine, Chapel Hill, NC, USA.
Sandy PreissRTI International, Research Triangle Park, NC, USA.
Daniel BrannockRTI International, Research Triangle Park, NC, USA.
John M BarattaDepartment of Physical Medicine and Rehabilitation, University of North Carolina at Chapel Hill School of Medicine, Chapel Hill, NC, USA.
Yun Jae YooDepartment of Hematology and Medical Oncology, Emory University, Atlanta, GA, USA.
Emily HadleyRTI International, Research Triangle Park, NC, USA.
Frank BlanceróJersey City, NJ, USA.
Robert ChewRTI International, Research Triangle Park, NC, USA.
Johanna LoombaIntegrated Translational Health Research Institute, University of Virginia, Charlottesville, VA, USA.
Abhishek BhatiaCarolina Health Informatics Program, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA.
Christopher G ChuteSection of Biomedical Informatics and Data Science, Johns Hopkins University, Baltimore, MD, USA.
Melissa HaendelDepartment of Genetics, University of North Carolina at Chapel Hill School of Medicine, Chapel Hill, NC, USA.
Richard MoffittDepartment of Hematology and Medical Oncology, Emory University, Atlanta, GA, USA.
Emily R PfaffDepartment of Medicine, University of North Carolina at Chapel Hill School of Medicine, Chapel Hill, NC, USA. Electronic address: epfaff@email.unc.edu.
N3C Consortium and the RECOVER EHR cohort

Funding

OTA-21-015A Post-Acute Sequelae of SARS-CoV-2 Infection Initiative: NYU Langone Health Clinical Science Core, Data Resource Core, and PASC Biorepository CoreOT2HL161847 · NHLBI · NEW YORK UNIVERSITY SCHOOL OF MEDICINE · PI GROSS, RACHEL SHARON, HORWITZ, LEORA · 2021 to 2025
$651.0M
Vanderbilt Institute for Clinical and Translational Research (VICTR) -Identifying correlates of functional immunity in SARS-CoV-2 convalescent plasmaUL1TR002243 · NCATS · VANDERBILT UNIVERSITY MEDICAL CENTER · PI Paul A. Harris, Wesley H Self · 2017 to 2026
$130.7M
University of Pittsburgh Clinical and Translational Science InstituteUL1TR001857 · NCATS · UNIVERSITY OF PITTSBURGH AT PITTSBURGH · PI REIS, STEVEN E · 2016 to 2025
$129.3M
UCLA Clinical Translational Science InstituteUL1TR001881 · NCATS · UNIVERSITY OF CALIFORNIA LOS ANGELES · PI ARLEEN F. BROWN, ARASH NAEIM · 2016 to 2026
$118.1M
Clinical and Translational Science InstituteUL1TR001872 · NCATS · UNIVERSITY OF CALIFORNIA, SAN FRANCISCO · PI COLLARD, HAROLD R, JACOBY, VANESSA · 2016 to 2025
$112.1M
Project-005UL1TR001445 · NCATS · NEW YORK UNIVERSITY SCHOOL OF MEDICINE · PI BREDELLA, MIRIAM ANTOINETTE, HOCHMAN, JUDITH S · 2015 to 2025
$103.5M
Yale Clinical and Translational Science Award (U Component)UL1TR001863 · NCATS · YALE UNIVERSITY · PI John H. Krystal, LUCILA OHNO-MACHADO · 2016 to 2026
$102.9M
Phenotypic Diversity in COVID-19UL1TR001878 · NCATS · UNIVERSITY OF PENNSYLVANIA · PI FITZGERALD, GARRET A · 2016 to 2025
$102.4M
Transform Dissemination and Implementation Science in CTSA ProgramsUL1TR002319 · NCATS · UNIVERSITY OF WASHINGTON · PI John K. Amory · 2017 to 2026
$100.0M
Clinical and Translational Science AwardUL1TR001873 · NCATS · COLUMBIA UNIVERSITY HEALTH SCIENCES · PI REILLY, MUREDACH P · 2016 to 2025
$99.0M
WU INSTITUTE OF CLINICAL AND TRANSLATIONAL SCIENCESUL1TR002345 · NCATS · WASHINGTON UNIVERSITY · PI William G. Powderly · 2017 to 2026
$97.8M
The Harvard Clinical and Translational Science CenterUL1TR002541 · NCATS · HARVARD MEDICAL SCHOOL · PI NADLER, LEE MARSHALL · 2018 to 2022
$93.0M
Gates Foundation INV-018455NCATS NIH HHS U24 TR002306NCATS NIH HHS UL1 TR000460NCATS NIH HHS UL1 TR001409NCATS NIH HHS UL1 TR001412NCATS NIH HHS UL1 TR001414NCATS NIH HHS UL1 TR001420NCATS NIH HHS UL1 TR001422NCATS NIH HHS UL1 TR001425NCATS NIH HHS UL1 TR001427NCATS NIH HHS UL1 TR001430NCATS NIH HHS UL1 TR001433NCATS NIH HHS UL1 TR001436NCATS NIH HHS UL1 TR001439NCATS NIH HHS UL1 TR001442NCATS NIH HHS UL1 TR001445NCATS NIH HHS UL1 TR001449NCATS NIH HHS UL1 TR001450NCATS NIH HHS UL1 TR001453NCATS NIH HHS UL1 TR001855NCATS NIH HHS UL1 TR001857NCATS NIH HHS UL1 TR001860NCATS NIH HHS UL1 TR001863NCATS NIH HHS UL1 TR001866NCATS NIH HHS UL1 TR001872NCATS NIH HHS UL1 TR001873NCATS NIH HHS UL1 TR001876NCATS NIH HHS UL1 TR001878NCATS NIH HHS UL1 TR001881NCATS NIH HHS UL1 TR001998NCATS NIH HHS UL1 TR002001NCATS NIH HHS UL1 TR002003NCATS NIH HHS UL1 TR002014NCATS NIH HHS UL1 TR002240NCATS NIH HHS UL1 TR002243NCATS NIH HHS UL1 TR002319NCATS NIH HHS UL1 TR002345NCATS NIH HHS UL1 TR002366NCATS NIH HHS UL1 TR002369NCATS NIH HHS UL1 TR002373NCATS NIH HHS UL1 TR002377NCATS NIH HHS UL1 TR002378NCATS NIH HHS UL1 TR002384NCATS NIH HHS UL1 TR002389NCATS NIH HHS UL1 TR002489NCATS NIH HHS UL1 TR002494NCATS NIH HHS UL1 TR002529NCATS NIH HHS UL1 TR002535NCATS NIH HHS UL1 TR002537NCATS NIH HHS UL1 TR002538NCATS NIH HHS UL1 TR002541NCATS NIH HHS UL1 TR002544NCATS NIH HHS UL1 TR002548NCATS NIH HHS UL1 TR002550NCATS NIH HHS UL1 TR002553NCATS NIH HHS UL1 TR002556NCATS NIH HHS UL1 TR002645NCATS NIH HHS UL1 TR002649NCATS NIH HHS UL1 TR002733NCATS NIH HHS UL1 TR002736NCATS NIH HHS UL1 TR003015NCATS NIH HHS UL1 TR003017NCATS NIH HHS UL1 TR003096NCATS NIH HHS UL1 TR003098NCATS NIH HHS UL1 TR003107NCATS NIH HHS UL1 TR003142NCATS NIH HHS UL1 TR003167NCATS NIH HHS UM1 TR004528NCATS NIH HHS UM1 TR004556NHLBI NIH HHS OT2 HL161847NIGMS NIH HHS U54 GM104938NIGMS NIH HHS U54 GM104940NIGMS NIH HHS U54 GM104941NIGMS NIH HHS U54 GM104942NIGMS NIH HHS U54 GM115371NIGMS NIH HHS U54 GM115428NIGMS NIH HHS U54 GM115458NIGMS NIH HHS U54 GM115516NIGMS NIH HHS U54 GM115677NIGMS NIH HHS U54 GM133807
6 · The paper itself

Abstract

backgroundIn 2021, we used the National COVID Cohort Collaborative (N3C) as part of the National Institutes of Health RECOVER Initiative to develop a machine learning pipeline to identify patients with a high probability of having post-acute sequelae of SARS-CoV-2 infection or long COVID. However, the increased home testing, missing documentation, and reinfections that characterise the pandemic beyond 2022 necessitated the re-engineering of our original model to account for these changes in the COVID-19 research landscape.

methodsTrained on 72 745 patient records (36 238 with long COVID and 36 507 with no evidence of long COVID), our updated XGBoost model gathered data for each patient in overlapping 100-day periods that progressed through time and issued a probability of long COVID for each 100-day period. We ran the model on patients in N3C (n=5 875 065) who met at least one of the following criteria from Jan 1, 2020, to June 22, 2023: a U07·1 (COVID-19) diagnosis code; a positive SARS-CoV-2 test; a U09·9 (post-acute sequelae of SARS-CoV-2 infection) diagnosis code; a prescription for nirmatrelvir-ritonavir or remdesivir; or an M35·81 (multisystem inflammatory syndrome in children [MIS-C]) diagnosis code. Each patient was given a model score that predicted long COVID status for each 100-day window in which they were aged ≥18 years. If a patient had known acute COVID-19 during any 100-day window (including reinfections), we censored the data from 7 days before the diagnosis or positive test date to 28 days after. We ran the model on controls selected from pre-2020 data to assess the likelihood of false positives.

findingsThe updated model had an area under the receiver operating characteristic curve of 0·90. Precision and recall could be adjusted according to a given use case, depending on whether greater sensitivity or specificity was warranted. Using our model, we estimate the overall prevalence of long COVID among the COVID-19 positive cohort within N3C repository to be 10.4%.

interpretationBy eschewing the COVID-19 index date as an anchor point for analysis, we can assess the probability of long COVID among patients who might have tested at home, or with suspected (but untested) cases of COVID-19, or multiple SARS-CoV-2 reinfections. We view this exercise as a model for maintaining and updating any machine learning pipeline used for clinical research and operations.

fundingNational Institutes of Health RECOVER Initiative.

Indexed as

COVID-19Machine LearningAdenosine MonophosphateAdultAlanineFemaleHumansMaleMiddle AgedSARS-CoV-2United StatesAdenosine MonophosphateAlanineremdesivir

Identifiers

PMID40858480
PMCPMC12439339

What OpenQuestion holds

Textmetadata
LicenceCC BY-NC-ND
Read underepoch 390

Registered trials

None linked

Read under generation 80e0d062 · epoch 390. Bibliography from PubMed, PubMed Central and OpenAlex; grants from NIH RePORTER; trial links from ClinicalTrials.gov; estimates, votes and beliefs from the OpenQuestion graph.