TY - JOUR
T1 - Use of Real-World EMR Data to Rapidly Evaluate Treatment Effects of Existing Drugs for Emerging Infectious Diseases
T2 - Remdesivir for COVID-19 Treatment as an Example
AU - Zhang, Chenguang
AU - Nigo, Masayuki
AU - Patel, Shivani
AU - Yu, Duo
AU - Septimus, Edward
AU - Wu, Hulin
N1 - Publisher Copyright:
© 2024, The Author(s) under exclusive licence to International Chinese Statistical Association.
PY - 2024
Y1 - 2024
N2 - For an emerging infectious disease such as 2019 coronavirus disease (COVID-19), initially there may not be any existing medication or treatment immediately available, which may result in high morbidity and mortality in a short time of period. In this case, it is urgent to quickly identify whether existing medications or treatments could be repurposed to treat the newly appeared disease before time-consuming randomized clinical trials (RCTs) can be done and new drugs can be developed. For example, when SARS-CoV-2 appeared in late 2019, clinicians started to use existing antiviral drugs, anti-inflammatory drugs, immune-based therapies and other types of medications to treat COVID-19 patients before any data or evidence was available to support the use of these medications for the new COVID-19 disease. Most of these medications have proven to be ineffective or only marginally effective to treat COVID-19 patients by more rigorous RCT or secondary data analyses later. We propose to use real-world electronic medical records (EMR) data to develop real-time treatment evaluation and monitoring systems to identify effective treatments or avoid ineffective treatments for emerging diseases in the future. In order to do this, first we have to deal with the challenges in processing and analyzing complex and noisy EMR data. In this paper, we outline these challenges and propose practical statistical methods and guidelines, which are derived from a project in evaluating anti-viral medication, remdesivir, for COVID-19 treatment based on a local healthcare EMR database.
AB - For an emerging infectious disease such as 2019 coronavirus disease (COVID-19), initially there may not be any existing medication or treatment immediately available, which may result in high morbidity and mortality in a short time of period. In this case, it is urgent to quickly identify whether existing medications or treatments could be repurposed to treat the newly appeared disease before time-consuming randomized clinical trials (RCTs) can be done and new drugs can be developed. For example, when SARS-CoV-2 appeared in late 2019, clinicians started to use existing antiviral drugs, anti-inflammatory drugs, immune-based therapies and other types of medications to treat COVID-19 patients before any data or evidence was available to support the use of these medications for the new COVID-19 disease. Most of these medications have proven to be ineffective or only marginally effective to treat COVID-19 patients by more rigorous RCT or secondary data analyses later. We propose to use real-world electronic medical records (EMR) data to develop real-time treatment evaluation and monitoring systems to identify effective treatments or avoid ineffective treatments for emerging diseases in the future. In order to do this, first we have to deal with the challenges in processing and analyzing complex and noisy EMR data. In this paper, we outline these challenges and propose practical statistical methods and guidelines, which are derived from a project in evaluating anti-viral medication, remdesivir, for COVID-19 treatment based on a local healthcare EMR database.
KW - Causal inference
KW - Competing risks
KW - EMR
KW - Propensity score
KW - Real-world data
KW - SARS-CoV-2
UR - http://www.scopus.com/inward/record.url?scp=85181221744&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85181221744&partnerID=8YFLogxK
U2 - 10.1007/s12561-023-09411-8
DO - 10.1007/s12561-023-09411-8
M3 - Article
AN - SCOPUS:85181221744
SN - 1867-1764
JO - Statistics in Biosciences
JF - Statistics in Biosciences
ER -