Ivan Lerner
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Ivan Lerner has a total of 103 co-authors. Here are the top co-authors by number of shared publications:
- Anita Burgun
Clinical Bioinformatics Laboratory, Imagine Institute, Université Paris Cité, INSERM UMR1163, 24 Boulevard du Montparnasse, Paris, F-75006, France, 33 668169844
9 shared publications - Antoine Neuraz
Department of biomedical informatics, Hôpital Necker-Enfants Malade, Assistance Publique Hôpitaux de Paris (APHP), University of Paris, Paris, France
9 shared publications - Bastien Rance
Hôpital Européen Georges Pompidou
9 shared publications - Nicolas Garcelon
Data Science Platform Imagine Institute INSERM U1163 UPC Paris France
7 shared publications - Arnaud Serret-Larmande
Sorbonne Univ, Ctr Rech Cordeliers, INSERM, Paris, France
4 shared publications - Laurent Chouchana
Pharmacovigilance Regional Centre, Pharmacology Department, Cochin Hospital, APHP UMR1343 INSERM Université Paris Cité Paris France
4 shared publications
Latest Publications
- Open Access
- 9 September 2026
- Published by Wiley in American Journal of Hematology
- Vol. 101 (12) , 2961-2975
The Presence and Percentage of Circulating Nucleated Red Blood Cells Reveal Distinct Characteristics in Adults With Sickle Cell Disease
Although circulating nucleated red blood cells (cNRBCs) have previously been detected in patients with sickle cell disease (SCD), their prevalence and clinical significance during steady‐state disease have not been established. We retrospectively analyzed 270 adults with an SS or Sβ 0 ‐thalassemia genotype. cNRBCs were counted during steady‐state disease. Patients were classified as high cNRBC+ (> 3 per 100 white blood cells (WBCs)), low cNRBC+ (1–3 per 100 WBCs), or cNRBC− (no cNRBCs). cNRBC+ and cNRBC− patients were compared with regard to clinical and laboratory data and in vitro erythroid differentiation from peripheral blood CD34+ cells. cNRBCs were detected in 65.6% of the patients and encompassed all terminal erythroid stages. High cNRBC+ patients were more likely to be receiving hydroxyurea ( p = 0.024). cNRBC+ patients exhibited a lower median [IQR] hemoglobin level (8.8 [7.8–10] and 8.4 [7.4–9.4] in the high and low cNRBC+ groups, respectively vs. 9.3 [8.3–9.9] g/dL in the cNRBC− group; p = 0.024). The cNRBC+ percentage was correlated negatively with the reticulocyte and WBC counts and positively with the mean corpuscular volume and the ferritin level. Clinically, a trend toward a greater likelihood of acute chest syndrome was observed in cNRBC+ patients. High cNRBC+ patients' cells exhibited accelerated erythroid differentiation in vitro, suggestive of dyserythropoiesis. cNRBCs constitute a common, easily measurable marker of a distinct hematological phenotype in adult patients with SCD: more severe anemia and signs of dyserythropoiesis. Prospective studies are needed to determine the value of the cNRBC percentage for predicting disease severity and guiding the treatment of dyserythropoiesis.
- 9 September 2026
- Published by Wiley in American Journal of Hematology
- Vol. 101 (12) , 2961-2975
Open AccessThe Presence and Percentage of Circulating Nucleated Red Blood Cells Reveal Distinct Characteristics in Adults With Sickle Cell Disease
Although circulating nucleated red blood cells (cNRBCs) have previously been detected in patients with sickle cell disease (SCD), their prevalence and clinical significance during steady‐state disease have not been established. We retrospectively analyzed 270 adults with an SS or Sβ 0 ‐thalassemia genotype. cNRBCs were counted during steady‐state disease. Patients were classified as high cNRBC+ (> 3 per 100 white blood cells (WBCs)), low cNRBC+ (1–3 per 100 WBCs), or cNRBC− (no cNRBCs). cNRBC+ and cNRBC− patients were compared with regard to clinical and laboratory data and in vitro erythroid differentiation from peripheral blood CD34+ cells. cNRBCs were detected in 65.6% of the patients and encompassed all terminal erythroid stages. High cNRBC+ patients were more likely to be receiving hydroxyurea ( p = 0.024). cNRBC+ patients exhibited a lower median [IQR] hemoglobin level (8.8 [7.8–10] and 8.4 [7.4–9.4] in the high and low cNRBC+ groups, respectively vs. 9.3 [8.3–9.9] g/dL in the cNRBC− group; p = 0.024). The cNRBC+ percentage was correlated negatively with the reticulocyte and WBC counts and positively with the mean corpuscular volume and the ferritin level. Clinically, a trend toward a greater likelihood of acute chest syndrome was observed in cNRBC+ patients. High cNRBC+ patients' cells exhibited accelerated erythroid differentiation in vitro, suggestive of dyserythropoiesis. cNRBCs constitute a common, easily measurable marker of a distinct hematological phenotype in adult patients with SCD: more severe anemia and signs of dyserythropoiesis. Prospective studies are needed to determine the value of the cNRBC percentage for predicting disease severity and guiding the treatment of dyserythropoiesis.
- 9 April 2026
- Published by Agence Bibliographique de l'Enseignement Supérieur
Sur la conception et l'apprentissage de structures invariantes dans les modèles de séries temporelles cliniques
Les dossiers patients informatisés (DPI) contiennent des données collectées dans les hôpitaux. Les DPI posent de multiples défis à l'apprentissage automatique. Nous nous intéressons aux aspects suivants : l'apprentissage à partir de séries temporelles échantillonnées de manière irrégulière et l'apprentissage de la structure de dépendances causale des modèles. Pour ce faire, nous nous sommes appuyés sur les processus gaussiens (GP) et des modèles causaux structurés (MCS). Notre première contribution a eu lieu dans le contexte de l'épidémie de COVID-19. Notre objectif était de générer des hypothèses pharmacologiques sans avoir besoin d'une connaissance a priori de la maladie. Nous avons développé le pipeline d'analyse Pharmacopeia-Wide Association Study (PharmWAS), qui recherche systématiquement des associations entre l'ensemble de la pharmacopée et un phénotype clinique. L'étude PharmWAS a mis en évidence la nécessité de développer des modèles d'apprentissage automatique adaptés qui tiennent compte de la nature dynamique des DPI et qui représentent de manière flexible la structure de dépendances au sein d'un graphe acyclique dirigé. Dans notre deuxième contribution, nous avons développé le modèle LPP-GP (latent patient pathways Gaussian process), un modèle interprétable et flexible pour l'apprentissage de dépendances entre les patients, et au sein de chaque patient entre les séries temporelles. Le modèle LPP-GP permet de découvrir des chemins latents dans une cohorte de patients. Le modèle LPP-GP permet un apprentissage par transfert interprétable et traçable entre les tâches et entre les patients. Enfin, constatant que l'ajout d'une simple contrainte sur les paramètres du modèle LPP-GP permettait de paramétrer un MCS dans le domaine des fréquences, nous avons développé le modèle StructGP, notre troisième contribution. Le modèle StructGP identifie un graphe de relations d'indépendance conditionnelle ordonnées entre des séries temporelles échantillonnées de manière irrégulière. Nous avons mis en oeuvre un algorithme d'apprentissage de la structure spectrale qui apprend le graphe à partir des observations et recouvre le véritable graphe avec une sensibilité élevée et une bonne précision.
- 9 April 2026
- Published by Agence Bibliographique de l'Enseignement Supérieur
Sur la conception et l'apprentissage de structures invariantes dans les modèles de séries temporelles cliniques
Les dossiers patients informatisés (DPI) contiennent des données collectées dans les hôpitaux. Les DPI posent de multiples défis à l'apprentissage automatique. Nous nous intéressons aux aspects suivants : l'apprentissage à partir de séries temporelles échantillonnées de manière irrégulière et l'apprentissage de la structure de dépendances causale des modèles. Pour ce faire, nous nous sommes appuyés sur les processus gaussiens (GP) et des modèles causaux structurés (MCS). Notre première contribution a eu lieu dans le contexte de l'épidémie de COVID-19. Notre objectif était de générer des hypothèses pharmacologiques sans avoir besoin d'une connaissance a priori de la maladie. Nous avons développé le pipeline d'analyse Pharmacopeia-Wide Association Study (PharmWAS), qui recherche systématiquement des associations entre l'ensemble de la pharmacopée et un phénotype clinique. L'étude PharmWAS a mis en évidence la nécessité de développer des modèles d'apprentissage automatique adaptés qui tiennent compte de la nature dynamique des DPI et qui représentent de manière flexible la structure de dépendances au sein d'un graphe acyclique dirigé. Dans notre deuxième contribution, nous avons développé le modèle LPP-GP (latent patient pathways Gaussian process), un modèle interprétable et flexible pour l'apprentissage de dépendances entre les patients, et au sein de chaque patient entre les séries temporelles. Le modèle LPP-GP permet de découvrir des chemins latents dans une cohorte de patients. Le modèle LPP-GP permet un apprentissage par transfert interprétable et traçable entre les tâches et entre les patients. Enfin, constatant que l'ajout d'une simple contrainte sur les paramètres du modèle LPP-GP permettait de paramétrer un MCS dans le domaine des fréquences, nous avons développé le modèle StructGP, notre troisième contribution. Le modèle StructGP identifie un graphe de relations d'indépendance conditionnelle ordonnées entre des séries temporelles échantillonnées de manière irrégulière. Nous avons mis en oeuvre un algorithme d'apprentissage de la structure spectrale qui apprend le graphe à partir des observations et recouvre le véritable graphe avec une sensibilité élevée et une bonne précision.
- Open Access
- 15 May 2025
- Published by Elsevier in Computer Methods and Programs in Biomedicine
- Vol. 268, 108857
SIMSAMU - A French medical dispatch dialog open dataset
- 15 May 2025
- Published by Elsevier in Computer Methods and Programs in Biomedicine
- Vol. 268, 108857
Open AccessSIMSAMU - A French medical dispatch dialog open dataset
- Open Access
- 27 September 2024
- Published by JMIR Publications Inc. in JMIR Public Health and Surveillance
- Vol. 10 (1) , e43563
Functional Impairment in Individuals Exposed to Violence Based on Electronical Forensic Medical Record Mining and Their Profile Identification: Controlled Observational Study
Background: Little is known about the functional consequences of violence when directly assessed as a primary outcome, and even less about how consistently these consequences are evaluated in a judicial context. The World Health Organization (WHO) highlighted the importance of a functional approach to health in 2001 with the release of the International Classification of Functioning, Disability, and Health (ICF). In most European countries, forensic physicians assess individuals exposed to violence to evaluate the outcomes of violence, providing certified medical evidence for magistrates’ sentencing decisions. This evaluation involves a mix of objective, subjective, and contextual elements, such as reported symptoms of fear, pain, and details of the assault. Quantifying these subjective elements with scales could enhance their interpretation and application in a judicial context. Objective: This study aims to (1) characterize and (2) assess 6 scales measuring subjective elements of functional impairment among individuals exposed to violence. Methods: We conducted a retrospective study that included individuals exposed to violence examined in a French department of forensic medicine over 12 months. A typology of violence encountered in medical settings was built based on the mining of electronic health records and the use of pattern recognition algorithms. The optimal number of violence types was determined using a robust and stable clustering approach, involving sample resampling and a multimetric scheme. Patients were then paired according to their homogeneous profiles, and the intra- and interrater reproducibility of the scales was evaluated. Results: All pain, fear, and life threat scales were significantly associated with higher functional impairment, suggesting that these measures contribute to the overall assessment of functional impairment. The intra- and interrater reproducibility of scales among similar situations of violence was measured, ranging from mild to good, with coefficients of concordance between 0.46-0.66 and 0.43-0.66, respectively. Individuals reporting intimate partner violence showed higher scores in both fear and perception of a life threat during the assault and medical interview, while individuals reporting battery by multiple unknown assailants presented higher scores only in perception of a life threat during the assault. We identified 5 remarkably stable profiles of situations of violence, consistent with clinical practice. Conclusions: Pain, fear, and life threat scales were related to functional impairment according to expert knowledge and demonstrated fair reproducibility under real-life conditions for similar situations of violence. Subjective elements related to functional impairment in individuals exposed to violence can be quantified using Likert scales during medical interviews.
- 27 September 2024
- Published by JMIR Publications Inc. in JMIR Public Health and Surveillance
- Vol. 10 (1) , e43563
Open AccessFunctional Impairment in Individuals Exposed to Violence Based on Electronical Forensic Medical Record Mining and Their Profile Identification: Controlled Observational Study
Background: Little is known about the functional consequences of violence when directly assessed as a primary outcome, and even less about how consistently these consequences are evaluated in a judicial context. The World Health Organization (WHO) highlighted the importance of a functional approach to health in 2001 with the release of the International Classification of Functioning, Disability, and Health (ICF). In most European countries, forensic physicians assess individuals exposed to violence to evaluate the outcomes of violence, providing certified medical evidence for magistrates’ sentencing decisions. This evaluation involves a mix of objective, subjective, and contextual elements, such as reported symptoms of fear, pain, and details of the assault. Quantifying these subjective elements with scales could enhance their interpretation and application in a judicial context. Objective: This study aims to (1) characterize and (2) assess 6 scales measuring subjective elements of functional impairment among individuals exposed to violence. Methods: We conducted a retrospective study that included individuals exposed to violence examined in a French department of forensic medicine over 12 months. A typology of violence encountered in medical settings was built based on the mining of electronic health records and the use of pattern recognition algorithms. The optimal number of violence types was determined using a robust and stable clustering approach, involving sample resampling and a multimetric scheme. Patients were then paired according to their homogeneous profiles, and the intra- and interrater reproducibility of the scales was evaluated. Results: All pain, fear, and life threat scales were significantly associated with higher functional impairment, suggesting that these measures contribute to the overall assessment of functional impairment. The intra- and interrater reproducibility of scales among similar situations of violence was measured, ranging from mild to good, with coefficients of concordance between 0.46-0.66 and 0.43-0.66, respectively. Individuals reporting intimate partner violence showed higher scores in both fear and perception of a life threat during the assault and medical interview, while individuals reporting battery by multiple unknown assailants presented higher scores only in perception of a life threat during the assault. We identified 5 remarkably stable profiles of situations of violence, consistent with clinical practice. Conclusions: Pain, fear, and life threat scales were related to functional impairment according to expert knowledge and demonstrated fair reproducibility under real-life conditions for similar situations of violence. Subjective elements related to functional impairment in individuals exposed to violence can be quantified using Likert scales during medical interviews.
- Open Access
- 15 October 2022
- Published by JMIR Publications Inc. in JMIR Preprints
Functional Impairment in Individuals Exposed to Violence Based on Electronical Forensic Medical Record Mining and Their Profile Identification: Controlled Observational Study (Preprint)
BACKGROUND Little is known about the functional consequences of violence when directly assessed as a primary outcome, and even less about how consistently these consequences are evaluated in a judicial context. The World Health Organization (WHO) highlighted the importance of a functional approach to health in 2001 with the release of the International Classification of Functioning, Disability, and Health (ICF). In most European countries, forensic physicians assess individuals exposed to violence to evaluate the outcomes of violence, providing certified medical evidence for magistrates’ sentencing decisions. This evaluation involves a mix of objective, subjective, and contextual elements, such as reported symptoms of fear, pain, and details of the assault. Quantifying these subjective elements with scales could enhance their interpretation and application in a judicial context. OBJECTIVE This study aims to (1) characterize and (2) assess 6 scales measuring subjective elements of functional impairment among individuals exposed to violence. METHODS We conducted a retrospective study that included individuals exposed to violence examined in a French department of forensic medicine over 12 months. A typology of violence encountered in medical settings was built based on the mining of electronic health records and the use of pattern recognition algorithms. The optimal number of violence types was determined using a robust and stable clustering approach, involving sample resampling and a multimetric scheme. Patients were then paired according to their homogeneous profiles, and the intra- and interrater reproducibility of the scales was evaluated. RESULTS All pain, fear, and life threat scales were significantly associated with higher functional impairment, suggesting that these measures contribute to the overall assessment of functional impairment. The intra- and interrater reproducibility of scales among similar situations of violence was measured, ranging from mild to good, with coefficients of concordance between 0.46-0.66 and 0.43-0.66, respectively. Individuals reporting intimate partner violence showed higher scores in both fear and perception of a life threat during the assault and medical interview, while individuals reporting battery by multiple unknown assailants presented higher scores only in perception of a life threat during the assault. We identified 5 remarkably stable profiles of situations of violence, consistent with clinical practice. CONCLUSIONS Pain, fear, and life threat scales were related to functional impairment according to expert knowledge and demonstrated fair reproducibility under real-life conditions for similar situations of violence. Subjective elements related to functional impairment in individuals exposed to violence can be quantified using Likert scales during medical interviews.
- 15 October 2022
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessFunctional Impairment in Individuals Exposed to Violence Based on Electronical Forensic Medical Record Mining and Their Profile Identification: Controlled Observational Study (Preprint)
BACKGROUND Little is known about the functional consequences of violence when directly assessed as a primary outcome, and even less about how consistently these consequences are evaluated in a judicial context. The World Health Organization (WHO) highlighted the importance of a functional approach to health in 2001 with the release of the International Classification of Functioning, Disability, and Health (ICF). In most European countries, forensic physicians assess individuals exposed to violence to evaluate the outcomes of violence, providing certified medical evidence for magistrates’ sentencing decisions. This evaluation involves a mix of objective, subjective, and contextual elements, such as reported symptoms of fear, pain, and details of the assault. Quantifying these subjective elements with scales could enhance their interpretation and application in a judicial context. OBJECTIVE This study aims to (1) characterize and (2) assess 6 scales measuring subjective elements of functional impairment among individuals exposed to violence. METHODS We conducted a retrospective study that included individuals exposed to violence examined in a French department of forensic medicine over 12 months. A typology of violence encountered in medical settings was built based on the mining of electronic health records and the use of pattern recognition algorithms. The optimal number of violence types was determined using a robust and stable clustering approach, involving sample resampling and a multimetric scheme. Patients were then paired according to their homogeneous profiles, and the intra- and interrater reproducibility of the scales was evaluated. RESULTS All pain, fear, and life threat scales were significantly associated with higher functional impairment, suggesting that these measures contribute to the overall assessment of functional impairment. The intra- and interrater reproducibility of scales among similar situations of violence was measured, ranging from mild to good, with coefficients of concordance between 0.46-0.66 and 0.43-0.66, respectively. Individuals reporting intimate partner violence showed higher scores in both fear and perception of a life threat during the assault and medical interview, while individuals reporting battery by multiple unknown assailants presented higher scores only in perception of a life threat during the assault. We identified 5 remarkably stable profiles of situations of violence, consistent with clinical practice. CONCLUSIONS Pain, fear, and life threat scales were related to functional impairment according to expert knowledge and demonstrated fair reproducibility under real-life conditions for similar situations of violence. Subjective elements related to functional impairment in individuals exposed to violence can be quantified using Likert scales during medical interviews.
- Open Access
- 12 April 2022
- Published by JMIR Publications Inc. in JMIR Medical Informatics
- Vol. 10 (4) , e38505-288
Correction: Mining Electronic Health Records for Drugs Associated With 28-day Mortality in COVID-19: Pharmacopoeia-wide Association Study (PharmWAS)
[This corrects the article DOI: 10.2196/35190.].
- 12 April 2022
- Published by JMIR Publications Inc. in JMIR Medical Informatics
- Vol. 10 (4) , e38505-288
Open AccessCorrection: Mining Electronic Health Records for Drugs Associated With 28-day Mortality in COVID-19: Pharmacopoeia-wide Association Study (PharmWAS)
[This corrects the article DOI: 10.2196/35190.].
- Open Access
- 5 April 2022
- Published by JMIR Publications Inc. in JMIR Preprints
Correction: Mining Electronic Health Records for Drugs Associated With 28-day Mortality in COVID-19: Pharmacopoeia-wide Association Study (PharmWAS) (Preprint)
Journal of Medical Internet Research - International Scientific Journal for Medical Research, Information and Communication on the Internet #Preprint #PeerReviewMe: Warning: This is a unreviewed preprint. Readers are warned that the document has not been peer-reviewed by expert/patient reviewers or an academic editor, may contain misleading claims, and is likely to undergo changes before final publication, if accepted, or may have been rejected/withdrawn. Readers with interest and expertise are encouraged to sign up as peer-reviewer, if the paper is within an open peer-review period. Please cite this preprint only for review purposes or for grant applications and CVs (if you are the author).
- 5 April 2022
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessCorrection: Mining Electronic Health Records for Drugs Associated With 28-day Mortality in COVID-19: Pharmacopoeia-wide Association Study (PharmWAS) (Preprint)
Journal of Medical Internet Research - International Scientific Journal for Medical Research, Information and Communication on the Internet #Preprint #PeerReviewMe: Warning: This is a unreviewed preprint. Readers are warned that the document has not been peer-reviewed by expert/patient reviewers or an academic editor, may contain misleading claims, and is likely to undergo changes before final publication, if accepted, or may have been rejected/withdrawn. Readers with interest and expertise are encouraged to sign up as peer-reviewer, if the paper is within an open peer-review period. Please cite this preprint only for review purposes or for grant applications and CVs (if you are the author).
- Open Access
- 30 March 2022
- Published by JMIR Publications Inc. in JMIR Medical Informatics
- Vol. 10 (3) , e35190
Mining Electronic Health Records for Drugs Associated With 28-day Mortality in COVID-19: Pharmacopoeia-wide Association Study (PharmWAS)
Background: Patients hospitalized for a given condition may be receiving other treatments for other contemporary conditions or comorbidities. The use of such observational clinical data for pharmacological hypothesis generation is appealing in the context of an emerging disease but particularly challenging due to the presence of drug indication bias. Objective: With this study, our main objective was the development and validation of a fully data-driven pipeline that would address this challenge. Our secondary objective was to generate pharmacological hypotheses in patients with COVID-19 and demonstrate the clinical relevance of the pipeline. Methods: We developed a pharmacopeia-wide association study (PharmWAS) pipeline inspired from the PheWAS methodology, which systematically screens for associations between the whole pharmacopeia and a clinical phenotype. First, a fully data-driven procedure based on adaptive least absolute shrinkage and selection operator (LASSO) determined drug-specific adjustment sets. Second, we computed several measures of association, including robust methods based on propensity scores (PSs) to control indication bias. Finally, we applied the Benjamini and Hochberg procedure of the false discovery rate (FDR). We applied this method in a multicenter retrospective cohort study using electronic medical records from 16 university hospitals of the Greater Paris area. We included all adult patients between 18 and 95 years old hospitalized in conventional wards for COVID-19 between February 1, 2020, and June 15, 2021. We investigated the association between drug prescription within 48 hours from admission and 28-day mortality. We validated our data-driven pipeline against a knowledge-based pipeline on 3 treatments of reference, for which experts agreed on the expected association with mortality. We then demonstrated its clinical relevance by screening all drugs prescribed in more than 100 patients to generate pharmacological hypotheses. Results: A total of 5783 patients were included in the analysis. The median age at admission was 69.2 (IQR 56.7-81.1) years, and 3390 (58.62%) of the patients were male. The performance of our automated pipeline was comparable or better for controlling bias than the knowledge-based adjustment set for 3 reference drugs: dexamethasone, phloroglucinol, and paracetamol. After correction for multiple testing, 4 drugs were associated with increased in-hospital mortality. Among these, diazepam and tramadol were the only ones not discarded by automated diagnostics, with adjusted odds ratios of 2.51 (95% CI 1.52-4.16, Q=.01) and 1.94 (95% CI 1.32-2.85, Q=.02), respectively. Conclusions: Our innovative approach proved useful in generating pharmacological hypotheses in an outbreak setting, without requiring a priori knowledge of the disease. Our systematic analysis of early prescribed treatments from patients hospitalized for COVID-19 showed that diazepam and tramadol are associated with increased 28-day mortality. Whether these drugs could worsen COVID-19 needs to be further assessed.
- 30 March 2022
- Published by JMIR Publications Inc. in JMIR Medical Informatics
- Vol. 10 (3) , e35190
Open AccessMining Electronic Health Records for Drugs Associated With 28-day Mortality in COVID-19: Pharmacopoeia-wide Association Study (PharmWAS)
Background: Patients hospitalized for a given condition may be receiving other treatments for other contemporary conditions or comorbidities. The use of such observational clinical data for pharmacological hypothesis generation is appealing in the context of an emerging disease but particularly challenging due to the presence of drug indication bias. Objective: With this study, our main objective was the development and validation of a fully data-driven pipeline that would address this challenge. Our secondary objective was to generate pharmacological hypotheses in patients with COVID-19 and demonstrate the clinical relevance of the pipeline. Methods: We developed a pharmacopeia-wide association study (PharmWAS) pipeline inspired from the PheWAS methodology, which systematically screens for associations between the whole pharmacopeia and a clinical phenotype. First, a fully data-driven procedure based on adaptive least absolute shrinkage and selection operator (LASSO) determined drug-specific adjustment sets. Second, we computed several measures of association, including robust methods based on propensity scores (PSs) to control indication bias. Finally, we applied the Benjamini and Hochberg procedure of the false discovery rate (FDR). We applied this method in a multicenter retrospective cohort study using electronic medical records from 16 university hospitals of the Greater Paris area. We included all adult patients between 18 and 95 years old hospitalized in conventional wards for COVID-19 between February 1, 2020, and June 15, 2021. We investigated the association between drug prescription within 48 hours from admission and 28-day mortality. We validated our data-driven pipeline against a knowledge-based pipeline on 3 treatments of reference, for which experts agreed on the expected association with mortality. We then demonstrated its clinical relevance by screening all drugs prescribed in more than 100 patients to generate pharmacological hypotheses. Results: A total of 5783 patients were included in the analysis. The median age at admission was 69.2 (IQR 56.7-81.1) years, and 3390 (58.62%) of the patients were male. The performance of our automated pipeline was comparable or better for controlling bias than the knowledge-based adjustment set for 3 reference drugs: dexamethasone, phloroglucinol, and paracetamol. After correction for multiple testing, 4 drugs were associated with increased in-hospital mortality. Among these, diazepam and tramadol were the only ones not discarded by automated diagnostics, with adjusted odds ratios of 2.51 (95% CI 1.52-4.16, Q=.01) and 1.94 (95% CI 1.32-2.85, Q=.02), respectively. Conclusions: Our innovative approach proved useful in generating pharmacological hypotheses in an outbreak setting, without requiring a priori knowledge of the disease. Our systematic analysis of early prescribed treatments from patients hospitalized for COVID-19 showed that diazepam and tramadol are associated with increased 28-day mortality. Whether these drugs could worsen COVID-19 needs to be further assessed.
- Open Access
- 4 January 2022
- Published by Elsevier in Biological Psychiatry Global Open Science
- Vol. 3 (1) , 56-67
Risk of Death in Individuals Hospitalized for COVID-19 With and Without Psychiatric Disorders: An Observational Multicenter Study in France
- 4 January 2022
- Published by Elsevier in Biological Psychiatry Global Open Science
- Vol. 3 (1) , 56-67
Open AccessRisk of Death in Individuals Hospitalized for COVID-19 With and Without Psychiatric Disorders: An Observational Multicenter Study in France
- Open Access
- 25 November 2021
- Published by JMIR Publications Inc. in JMIR Preprints
Mining Electronic Health Records for Drugs Associated With 28-day Mortality in COVID-19: Pharmacopoeia-wide Association Study (PharmWAS) (Preprint)
BACKGROUND Patients hospitalized for a given condition may be receiving other treatments for other contemporary conditions or comorbidities. The use of such observational clinical data for pharmacological hypothesis generation is appealing in the context of an emerging disease but particularly challenging due to the presence of drug indication bias. OBJECTIVE With this study, our main objective was the development and validation of a fully data-driven pipeline that would address this challenge. Our secondary objective was to generate pharmacological hypotheses in patients with COVID-19 and demonstrate the clinical relevance of the pipeline. METHODS We developed a pharmacopeia-wide association study (PharmWAS) pipeline inspired from the PheWAS methodology, which systematically screens for associations between the whole pharmacopeia and a clinical phenotype. First, a fully data-driven procedure based on adaptive least absolute shrinkage and selection operator (LASSO) determined drug-specific adjustment sets. Second, we computed several measures of association, including robust methods based on propensity scores (PSs) to control indication bias. Finally, we applied the Benjamini and Hochberg procedure of the false discovery rate (FDR). We applied this method in a multicenter retrospective cohort study using electronic medical records from 16 university hospitals of the Greater Paris area. We included all adult patients between 18 and 95 years old hospitalized in conventional wards for COVID-19 between February 1, 2020, and June 15, 2021. We investigated the association between drug prescription within 48 hours from admission and 28-day mortality. We validated our data-driven pipeline against a knowledge-based pipeline on 3 treatments of reference, for which experts agreed on the expected association with mortality. We then demonstrated its clinical relevance by screening all drugs prescribed in more than 100 patients to generate pharmacological hypotheses. RESULTS A total of 5783 patients were included in the analysis. The median age at admission was 69.2 (IQR 56.7-81.1) years, and 3390 (58.62%) of the patients were male. The performance of our automated pipeline was comparable or better for controlling bias than the knowledge-based adjustment set for 3 reference drugs: dexamethasone, phloroglucinol, and paracetamol. After correction for multiple testing, 4 drugs were associated with increased in-hospital mortality. Among these, diazepam and tramadol were the only ones not discarded by automated diagnostics, with adjusted odds ratios of 2.51 (95% CI 1.52-4.16, Q=.1) and 1.94 (95% CI 1.32-2.85, Q=.02), respectively. CONCLUSIONS Our innovative approach proved useful in generating pharmacological hypotheses in an outbreak setting, without requiring a priori knowledge of the disease. Our systematic analysis of early prescribed treatments from patients hospitalized for COVID-19 showed that diazepam and tramadol are associated with increased 28-day mortality. Whether these drugs could worsen COVID-19 needs to be further assessed.
- 25 November 2021
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessMining Electronic Health Records for Drugs Associated With 28-day Mortality in COVID-19: Pharmacopoeia-wide Association Study (PharmWAS) (Preprint)
BACKGROUND Patients hospitalized for a given condition may be receiving other treatments for other contemporary conditions or comorbidities. The use of such observational clinical data for pharmacological hypothesis generation is appealing in the context of an emerging disease but particularly challenging due to the presence of drug indication bias. OBJECTIVE With this study, our main objective was the development and validation of a fully data-driven pipeline that would address this challenge. Our secondary objective was to generate pharmacological hypotheses in patients with COVID-19 and demonstrate the clinical relevance of the pipeline. METHODS We developed a pharmacopeia-wide association study (PharmWAS) pipeline inspired from the PheWAS methodology, which systematically screens for associations between the whole pharmacopeia and a clinical phenotype. First, a fully data-driven procedure based on adaptive least absolute shrinkage and selection operator (LASSO) determined drug-specific adjustment sets. Second, we computed several measures of association, including robust methods based on propensity scores (PSs) to control indication bias. Finally, we applied the Benjamini and Hochberg procedure of the false discovery rate (FDR). We applied this method in a multicenter retrospective cohort study using electronic medical records from 16 university hospitals of the Greater Paris area. We included all adult patients between 18 and 95 years old hospitalized in conventional wards for COVID-19 between February 1, 2020, and June 15, 2021. We investigated the association between drug prescription within 48 hours from admission and 28-day mortality. We validated our data-driven pipeline against a knowledge-based pipeline on 3 treatments of reference, for which experts agreed on the expected association with mortality. We then demonstrated its clinical relevance by screening all drugs prescribed in more than 100 patients to generate pharmacological hypotheses. RESULTS A total of 5783 patients were included in the analysis. The median age at admission was 69.2 (IQR 56.7-81.1) years, and 3390 (58.62%) of the patients were male. The performance of our automated pipeline was comparable or better for controlling bias than the knowledge-based adjustment set for 3 reference drugs: dexamethasone, phloroglucinol, and paracetamol. After correction for multiple testing, 4 drugs were associated with increased in-hospital mortality. Among these, diazepam and tramadol were the only ones not discarded by automated diagnostics, with adjusted odds ratios of 2.51 (95% CI 1.52-4.16, Q=.1) and 1.94 (95% CI 1.32-2.85, Q=.02), respectively. CONCLUSIONS Our innovative approach proved useful in generating pharmacological hypotheses in an outbreak setting, without requiring a priori knowledge of the disease. Our systematic analysis of early prescribed treatments from patients hospitalized for COVID-19 showed that diazepam and tramadol are associated with increased 28-day mortality. Whether these drugs could worsen COVID-19 needs to be further assessed.