Performance drift in a mortality prediction algorithm among patients with cancer during the SARS-CoV-2 pandemic.

Autor:	Parikh RB; Department of Medicine, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania, USA.; Department of Medical Ethics and Health Policy, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania, USA.; Corporal Michael J. Crescenz VA Medical Center, Philadelphia, Pennsylvania, USA., Zhang Y; Department of Medical Ethics and Health Policy, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania, USA., Kolla L; Department of Medicine, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania, USA.; Department of Biostatistics, Epidemiology and Informatics, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania, USA., Chivers C; Department of Medicine, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania, USA., Courtright KR; Department of Medicine, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania, USA., Zhu J; Department of Medical Ethics and Health Policy, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania, USA., Navathe AS; Department of Medicine, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania, USA.; Department of Medical Ethics and Health Policy, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania, USA.; Corporal Michael J. Crescenz VA Medical Center, Philadelphia, Pennsylvania, USA., Chen J; Department of Biostatistics, Epidemiology and Informatics, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania, USA.
Jazyk:	angličtina
Zdroj:	Journal of the American Medical Informatics Association : JAMIA [J Am Med Inform Assoc] 2023 Jan 18; Vol. 30 (2), pp. 348-354.
DOI:	10.1093/jamia/ocac221
Abstrakt:	Sudden changes in health care utilization during the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pandemic may have impacted the performance of clinical predictive models that were trained prior to the pandemic. In this study, we evaluated the performance over time of a machine learning, electronic health record-based mortality prediction algorithm currently used in clinical practice to identify patients with cancer who may benefit from early advance care planning conversations. We show that during the pandemic period, algorithm identification of high-risk patients had a substantial and sustained decline. Decreases in laboratory utilization during the peak of the pandemic may have contributed to drift. Calibration and overall discrimination did not markedly decline during the pandemic. This argues for careful attention to the performance and retraining of predictive algorithms that use inputs from the pandemic period. (© The Author(s) 2022. Published by Oxford University Press on behalf of the American Medical Informatics Association. All rights reserved. For permissions, please email: journals.permissions@oup.com.)
Databáze:	MEDLINE
Externí odkaz:	Zobrazit plný text záznamu