How to Identify Potential Candidates for HIV Pre-Exposure Prophylaxis: An AI Algorithm Reusing Real-World Hospital Data. Duthe, J., Bouzille, G., Sylvestre, E., Chazard, E., Arvieux, C., & Cuggia, M. Studies in Health Technology and Informatics, 281:714–718, May, 2021. doi abstract bibtex HIV Pre-Exposure Prophylaxis (PrEP) is effective in Men who have Sex with Men (MSM), and is reimbursed by the social security in France. Yet, PrEP is underused due to the difficulty to identify people at risk of HIV infection outside the "sexual health" care path. We developed and validated an automated algorithm that re-uses Electronic Health Record (EHR) data available in eHOP, the Clinical Data Warehouse of Rennes University Hospital (France). Using machine learning methods, we developed five models to predict incident HIV infections with 162 variables that might be exploited to predict HIV risk using EHR data. We divided patients aged 18 or more having at least one hospital admission between 2013 and 2019 in two groups: cases (patients with known HIV infection in the study period) and controls (patients without known HIV infection and no PrEP in the study period, but with at least one HIV risk factor). Among the 624,708 admissions, we selected 156 cases (incident HIV infection) and 761 controls. The best performing model for identifying incident HIV infections was the combined model (LASSO, Random Forest, and Generalized Linear Model): AUC = 0.88 (95% CI: 0.8143-0.9619), specificity = 0.887, and sensitivity = 0.733 using the test dataset. The algorithm seems to efficiently identify patients at risk of HIV infection.
@article{duthe_how_2021,
title = {How to {Identify} {Potential} {Candidates} for {HIV} {Pre}-{Exposure} {Prophylaxis}: {An} {AI} {Algorithm} {Reusing} {Real}-{World} {Hospital} {Data}},
volume = {281},
issn = {1879-8365},
shorttitle = {How to {Identify} {Potential} {Candidates} for {HIV} {Pre}-{Exposure} {Prophylaxis}},
doi = {10.3233/SHTI210265},
abstract = {HIV Pre-Exposure Prophylaxis (PrEP) is effective in Men who have Sex with Men (MSM), and is reimbursed by the social security in France. Yet, PrEP is underused due to the difficulty to identify people at risk of HIV infection outside the "sexual health" care path. We developed and validated an automated algorithm that re-uses Electronic Health Record (EHR) data available in eHOP, the Clinical Data Warehouse of Rennes University Hospital (France). Using machine learning methods, we developed five models to predict incident HIV infections with 162 variables that might be exploited to predict HIV risk using EHR data. We divided patients aged 18 or more having at least one hospital admission between 2013 and 2019 in two groups: cases (patients with known HIV infection in the study period) and controls (patients without known HIV infection and no PrEP in the study period, but with at least one HIV risk factor). Among the 624,708 admissions, we selected 156 cases (incident HIV infection) and 761 controls. The best performing model for identifying incident HIV infections was the combined model (LASSO, Random Forest, and Generalized Linear Model): AUC = 0.88 (95\% CI: 0.8143-0.9619), specificity = 0.887, and sensitivity = 0.733 using the test dataset. The algorithm seems to efficiently identify patients at risk of HIV infection.},
language = {eng},
journal = {Studies in Health Technology and Informatics},
author = {Duthe, Jean-Charles and Bouzille, Guillaume and Sylvestre, Emmanuelle and Chazard, Emmanuel and Arvieux, Cedric and Cuggia, Marc},
month = may,
year = {2021},
pmid = {34042669},
keywords = {Algorithms, Anti-HIV Agents, France, HIV Infections, HIV prevention, Homosexuality, Male, Hospitals, Humans, Male, Pre-Exposure Prophylaxis, Pre-exposure prophylaxis (PrEP), Sexual and Gender Minorities, clinical informatics, machine learning, predictive analytics, risk reduction practices, sexual health},
pages = {714--718},
}
Downloads: 0
{"_id":"pWaJS94WHFpXWDdch","bibbaseid":"duthe-bouzille-sylvestre-chazard-arvieux-cuggia-howtoidentifypotentialcandidatesforhivpreexposureprophylaxisanaialgorithmreusingrealworldhospitaldata-2021","author_short":["Duthe, J.","Bouzille, G.","Sylvestre, E.","Chazard, E.","Arvieux, C.","Cuggia, M."],"bibdata":{"bibtype":"article","type":"article","title":"How to Identify Potential Candidates for HIV Pre-Exposure Prophylaxis: An AI Algorithm Reusing Real-World Hospital Data","volume":"281","issn":"1879-8365","shorttitle":"How to Identify Potential Candidates for HIV Pre-Exposure Prophylaxis","doi":"10.3233/SHTI210265","abstract":"HIV Pre-Exposure Prophylaxis (PrEP) is effective in Men who have Sex with Men (MSM), and is reimbursed by the social security in France. Yet, PrEP is underused due to the difficulty to identify people at risk of HIV infection outside the \"sexual health\" care path. We developed and validated an automated algorithm that re-uses Electronic Health Record (EHR) data available in eHOP, the Clinical Data Warehouse of Rennes University Hospital (France). Using machine learning methods, we developed five models to predict incident HIV infections with 162 variables that might be exploited to predict HIV risk using EHR data. We divided patients aged 18 or more having at least one hospital admission between 2013 and 2019 in two groups: cases (patients with known HIV infection in the study period) and controls (patients without known HIV infection and no PrEP in the study period, but with at least one HIV risk factor). Among the 624,708 admissions, we selected 156 cases (incident HIV infection) and 761 controls. The best performing model for identifying incident HIV infections was the combined model (LASSO, Random Forest, and Generalized Linear Model): AUC = 0.88 (95% CI: 0.8143-0.9619), specificity = 0.887, and sensitivity = 0.733 using the test dataset. The algorithm seems to efficiently identify patients at risk of HIV infection.","language":"eng","journal":"Studies in Health Technology and Informatics","author":[{"propositions":[],"lastnames":["Duthe"],"firstnames":["Jean-Charles"],"suffixes":[]},{"propositions":[],"lastnames":["Bouzille"],"firstnames":["Guillaume"],"suffixes":[]},{"propositions":[],"lastnames":["Sylvestre"],"firstnames":["Emmanuelle"],"suffixes":[]},{"propositions":[],"lastnames":["Chazard"],"firstnames":["Emmanuel"],"suffixes":[]},{"propositions":[],"lastnames":["Arvieux"],"firstnames":["Cedric"],"suffixes":[]},{"propositions":[],"lastnames":["Cuggia"],"firstnames":["Marc"],"suffixes":[]}],"month":"May","year":"2021","pmid":"34042669","keywords":"Algorithms, Anti-HIV Agents, France, HIV Infections, HIV prevention, Homosexuality, Male, Hospitals, Humans, Male, Pre-Exposure Prophylaxis, Pre-exposure prophylaxis (PrEP), Sexual and Gender Minorities, clinical informatics, machine learning, predictive analytics, risk reduction practices, sexual health","pages":"714–718","bibtex":"@article{duthe_how_2021,\n\ttitle = {How to {Identify} {Potential} {Candidates} for {HIV} {Pre}-{Exposure} {Prophylaxis}: {An} {AI} {Algorithm} {Reusing} {Real}-{World} {Hospital} {Data}},\n\tvolume = {281},\n\tissn = {1879-8365},\n\tshorttitle = {How to {Identify} {Potential} {Candidates} for {HIV} {Pre}-{Exposure} {Prophylaxis}},\n\tdoi = {10.3233/SHTI210265},\n\tabstract = {HIV Pre-Exposure Prophylaxis (PrEP) is effective in Men who have Sex with Men (MSM), and is reimbursed by the social security in France. Yet, PrEP is underused due to the difficulty to identify people at risk of HIV infection outside the \"sexual health\" care path. We developed and validated an automated algorithm that re-uses Electronic Health Record (EHR) data available in eHOP, the Clinical Data Warehouse of Rennes University Hospital (France). Using machine learning methods, we developed five models to predict incident HIV infections with 162 variables that might be exploited to predict HIV risk using EHR data. We divided patients aged 18 or more having at least one hospital admission between 2013 and 2019 in two groups: cases (patients with known HIV infection in the study period) and controls (patients without known HIV infection and no PrEP in the study period, but with at least one HIV risk factor). Among the 624,708 admissions, we selected 156 cases (incident HIV infection) and 761 controls. The best performing model for identifying incident HIV infections was the combined model (LASSO, Random Forest, and Generalized Linear Model): AUC = 0.88 (95\\% CI: 0.8143-0.9619), specificity = 0.887, and sensitivity = 0.733 using the test dataset. The algorithm seems to efficiently identify patients at risk of HIV infection.},\n\tlanguage = {eng},\n\tjournal = {Studies in Health Technology and Informatics},\n\tauthor = {Duthe, Jean-Charles and Bouzille, Guillaume and Sylvestre, Emmanuelle and Chazard, Emmanuel and Arvieux, Cedric and Cuggia, Marc},\n\tmonth = may,\n\tyear = {2021},\n\tpmid = {34042669},\n\tkeywords = {Algorithms, Anti-HIV Agents, France, HIV Infections, HIV prevention, Homosexuality, Male, Hospitals, Humans, Male, Pre-Exposure Prophylaxis, Pre-exposure prophylaxis (PrEP), Sexual and Gender Minorities, clinical informatics, machine learning, predictive analytics, risk reduction practices, sexual health},\n\tpages = {714--718},\n}\n\n","author_short":["Duthe, J.","Bouzille, G.","Sylvestre, E.","Chazard, E.","Arvieux, C.","Cuggia, M."],"key":"duthe_how_2021","id":"duthe_how_2021","bibbaseid":"duthe-bouzille-sylvestre-chazard-arvieux-cuggia-howtoidentifypotentialcandidatesforhivpreexposureprophylaxisanaialgorithmreusingrealworldhospitaldata-2021","role":"author","urls":{},"keyword":["Algorithms","Anti-HIV Agents","France","HIV Infections","HIV prevention","Homosexuality","Male","Hospitals","Humans","Male","Pre-Exposure Prophylaxis","Pre-exposure prophylaxis (PrEP)","Sexual and Gender Minorities","clinical informatics","machine learning","predictive analytics","risk reduction practices","sexual health"],"metadata":{"authorlinks":{}}},"bibtype":"article","biburl":"https://api.zotero.org/users/1597782/collections/MSB7W4UM/items?key=gxIPM4PJtMVcB8OpssCWodtP&format=bibtex&limit=100&start=0&sort=date","dataSources":["doevpoZ8x7wJceFTM"],"keywords":["algorithms","anti-hiv agents","france","hiv infections","hiv prevention","homosexuality","male","hospitals","humans","male","pre-exposure prophylaxis","pre-exposure prophylaxis (prep)","sexual and gender minorities","clinical informatics","machine learning","predictive analytics","risk reduction practices","sexual health"],"search_terms":["identify","potential","candidates","hiv","pre","exposure","prophylaxis","algorithm","reusing","real","world","hospital","data","duthe","bouzille","sylvestre","chazard","arvieux","cuggia"],"title":"How to Identify Potential Candidates for HIV Pre-Exposure Prophylaxis: An AI Algorithm Reusing Real-World Hospital Data","year":2021}