A latent transfer learning method for estimating hospital-specific post-acute healthcare demands following SARS-CoV-2 infection

Qiong Wu; Nathan M Pajor; Yiwen Lu; Charles J Wolock; Jiayi Tong; Vitaly Lorman; Kevin B Johnson; Jason H Moore; Christopher B Forrest; David A Asch; Yong Chen

doi:10.1016/j.patter.2024.101079

A latent transfer learning method for estimating hospital-specific post-acute healthcare demands following SARS-CoV-2 infection

Patterns (N Y). 2024 Oct 24;5(11):101079. doi: 10.1016/j.patter.2024.101079. eCollection 2024 Nov 8.

Authors

Qiong Wu^{1

2

3}, Nathan M Pajor⁴, Yiwen Lu^{2

5}, Charles J Wolock^{2

3}, Jiayi Tong^{2

3

6}, Vitaly Lorman⁷, Kevin B Johnson^{2

8}, Jason H Moore⁹, Christopher B Forrest⁷, David A Asch^{8

10}, Yong Chen^{2

3

5

8

11

12}

Affiliations

¹ Department of Biostatistics and Health Data Science, University of Pittsburgh, Pittsburgh, PA 15261, USA.
² Department of Biostatistics, Epidemiology, and Informatics, University of Pennsylvania Perelman School of Medicine, Philadelphia, PA 19104, USA.
³ The Center for Health AI and Synthesis of Evidence (CHASE), University of Pennsylvania, Philadelphia, PA 19104, USA.
⁴ Division of Pulmonary Medicine, Cincinnati Children's Hospital Medical Center and University of Cincinnati College of Medicine, Cincinnati, OH 45229, USA.
⁵ The Graduate Group in Applied Mathematics and Computational Science, School of Arts and Sciences, University of Pennsylvania, Philadelphia, PA 19104, USA.
⁶ Department of Biostatistics, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD 21205, USA.
⁷ Applied Clinical Research Center, Children's Hospital of Philadelphia, Philadelphia, PA 19104, USA.
⁸ Leonard Davis Institute for Health Economics, University of Pennsylvania, Philadelphia, PA 19104, USA.
⁹ Department of Computational Biomedicine, Cedars-Sinai Medical Center, Los Angeles, CA 90048, USA.
¹⁰ Division of General Internal Medicine, University of Pennsylvania Perelman School of Medicine, Philadelphia, PA 19104, USA.
¹¹ Penn Medicine Center for Evidence-based Practice (CEP), Philadelphia, PA 19104, USA.
¹² Penn Institute for Biomedical Informatics (IBI), Philadelphia, PA 19104, USA.

Abstract

The long-term complications of COVID-19, known as the post-acute sequelae of SARS-CoV-2 infection (PASC), significantly burden healthcare resources. Quantifying the demand for post-acute healthcare is essential for understanding patients' needs and optimizing the allocation of valuable medical resources for disease management. Driven by this need, we developed a heterogeneous latent transfer learning framework (Latent-TL) to generate critical insights for individual health systems in a distributed research network. Latent-TL enhances learning in a specific health system by borrowing information from all other health systems in the network in a data-driven fashion. By identifying subpopulations with varying healthcare needs, our Latent-TL framework can provide more effective guidance for decision-making. Applying Latent-TL to electronic health record (EHR) data from eight health systems in PEDSnet, a national learning health system in the US, revealed four distinct patient subpopulations with heterogeneous post-acute healthcare demands following COVID-19 infections, varying across subpopulations and hospitals.

Keywords: COVID-19; Long COVID; electronic health records; healthcare utilization; learning health system; real-world data; transfer learning.