Julie Josse
Research Subjects
Julie Josse has a total of 98 co-authors. Here are the top co-authors by number of shared publications:
François Husson
IRMAR, Agrocampus Ouest, CNRS, Rennes, France
11 shared publications- François Husson
Applied Mathematics Department Agrocampus Ouest 65 rue de Saint-Brieuc F-35042 RENNES Cedex France
3 shared publications - Gaël Varoquaux
Soda, Inria Saclay, Palaiseau, France
3 shared publications Stefan Wager
Ecole Polytechnique -INRIA University of Geneva Stanford University
3 shared publications- Vincent Audigier
Applied Mathematics Department Agrocampus Ouest 65 rue de Saint-Brieuc F-35042 RENNES Cedex France
3 shared publications - Alexandre Perez-Lebel
Soda project team Inria Saclay Palaiseau France
2 shared publications
Latest Publications
- Open Access
- 29 October 2026
- Published by MIT Press in Harvard Data Science Review
- Vol. 8 (4)
Causal Meta-Analysis: Rethinking the Foundations of Evidence-Based Medicine
Meta-analysis, by synthesizing effect estimates from multiple studies conducted in diverse settings, stands at the top of the evidence hierarchy in clinical research. Yet, conventional approaches based on fixed- or random-effects models lack a causal framework, which may limit their interpretability and utility for public policy. Incorporating causal inference reframes meta-analysis as the estimation of well-defined causal effects on clearly specified populations, enabling a principled approach to handling study heterogeneity. We show that classical meta-analysis estimators have a clear causal interpretation when effects are measured as risk differences. However, this breaks down for nonlinear measures like the risk ratio and odds ratio. To address this, we introduce novel causal aggregation formulas that remain compatible with standard meta-analysis practices and do not require access to individual-level data. To evaluate real-world impact, we apply both classical and causal meta-analysis methods to 500 published meta-analyses. While the conclusions often align, we unveil cases where conventional methods may suggest a treatment is beneficial when, under a causal lens, it is in fact harmful.
- 29 October 2026
- Published by MIT Press in Harvard Data Science Review
- Vol. 8 (4)
Open AccessCausal Meta-Analysis: Rethinking the Foundations of Evidence-Based Medicine
Meta-analysis, by synthesizing effect estimates from multiple studies conducted in diverse settings, stands at the top of the evidence hierarchy in clinical research. Yet, conventional approaches based on fixed- or random-effects models lack a causal framework, which may limit their interpretability and utility for public policy. Incorporating causal inference reframes meta-analysis as the estimation of well-defined causal effects on clearly specified populations, enabling a principled approach to handling study heterogeneity. We show that classical meta-analysis estimators have a clear causal interpretation when effects are measured as risk differences. However, this breaks down for nonlinear measures like the risk ratio and odds ratio. To address this, we introduce novel causal aggregation formulas that remain compatible with standard meta-analysis practices and do not require access to individual-level data. To evaluate real-world impact, we apply both classical and causal meta-analysis methods to 500 published meta-analyses. While the conclusions often align, we unveil cases where conventional methods may suggest a treatment is beneficial when, under a causal lens, it is in fact harmful.
- Open Access
- 19 February 2026
- Published by Elsevier in SSRN Electronic Journal
Real-World Effect of GLP-1 Receptor Agonist Initiation on Body Composition and Blood Pressure: A Distributional Causal Inference Analysis of Connected Device Data
Background: Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) are established treatments for obesity, yet real-world evidence regarding the physiological quality of weight loss, specifically the preservation of lean mass versus fat mass, remains limited outside of randomised clinical trials. Furthermore, the temporal dynamics of body composition and blood pressure changes in routine practice are under-characterised.Methods: We conducted a retrospective observational cohort study of adult users of Withings connected scales and blood pressure monitors in the USA. We compared a cohort of self-reported GLP-1 RA users to non-users, employing a distributional causal inference framework (ADD-MALTS). This approach matched users based on the full distribution of their pre-treatment measurements (capturing stability and volatility) rather than scalar averages. We estimated the distributional Average Treatment Effect on the Treated and scalar median differences during a one-year efficacy window (months 10–14 post-initiation).Findings: The analysis included 2,031 users for body composition (396 exposed) and 672 for blood pressure (148 exposed). In the one-year efficacy window, exposed users achieved a median total weight reduction of 9·8% (95% CI 9·8–9·9) relative to the matched unexposed group. This reduction was driven predominantly by device-estimated fat mass loss (median 7·5% [7·5–7·8]), with a proportionally smaller reduction in device-estimated muscle mass (median 2·3% [2·1–2·4]). Significant reductions were also observed in systolic (median 2·5 mm Hg [1·5–2·9]) and diastolic blood pressure (median 1·5 mm Hg [1·2–1·9]).Interpretation: In this real-world cohort, GLP-1 RA therapy was associated with clinically meaningful weight loss characterised by a favourable ratio of fat-to-lean mass reduction, alongside sustained blood pressure improvements. These findings suggest relative muscle preservation, although reliance on bio-impedance precludes definitive conclusions regarding sarcopenia. Distributional causal inference successfully leveraged wearable sensor data to estimate effectiveness across a heterogeneous therapeutic class.
- 19 February 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessReal-World Effect of GLP-1 Receptor Agonist Initiation on Body Composition and Blood Pressure: A Distributional Causal Inference Analysis of Connected Device Data
Background: Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) are established treatments for obesity, yet real-world evidence regarding the physiological quality of weight loss, specifically the preservation of lean mass versus fat mass, remains limited outside of randomised clinical trials. Furthermore, the temporal dynamics of body composition and blood pressure changes in routine practice are under-characterised.Methods: We conducted a retrospective observational cohort study of adult users of Withings connected scales and blood pressure monitors in the USA. We compared a cohort of self-reported GLP-1 RA users to non-users, employing a distributional causal inference framework (ADD-MALTS). This approach matched users based on the full distribution of their pre-treatment measurements (capturing stability and volatility) rather than scalar averages. We estimated the distributional Average Treatment Effect on the Treated and scalar median differences during a one-year efficacy window (months 10–14 post-initiation).Findings: The analysis included 2,031 users for body composition (396 exposed) and 672 for blood pressure (148 exposed). In the one-year efficacy window, exposed users achieved a median total weight reduction of 9·8% (95% CI 9·8–9·9) relative to the matched unexposed group. This reduction was driven predominantly by device-estimated fat mass loss (median 7·5% [7·5–7·8]), with a proportionally smaller reduction in device-estimated muscle mass (median 2·3% [2·1–2·4]). Significant reductions were also observed in systolic (median 2·5 mm Hg [1·5–2·9]) and diastolic blood pressure (median 1·5 mm Hg [1·2–1·9]).Interpretation: In this real-world cohort, GLP-1 RA therapy was associated with clinically meaningful weight loss characterised by a favourable ratio of fat-to-lean mass reduction, alongside sustained blood pressure improvements. These findings suggest relative muscle preservation, although reliance on bio-impedance precludes definitive conclusions regarding sarcopenia. Distributional causal inference successfully leveraged wearable sensor data to estimate effectiveness across a heterogeneous therapeutic class.
- Top Cited PapersOpen Access
- 15 April 2022
- Published by Oxford University Press (OUP) in GigaScience
- Vol. 11
Benchmarking missing-values approaches for predictive models on health databases
Background: As databases grow larger, it becomes harder to fully control their collection, and they frequently come with missing values. These large databases are well suited to train machine learning models, e.g., for forecasting or to extract biomarkers in biomedical settings. Such predictive approaches can use discriminative—rather than generative—modeling and thus open the door to new missing-values strategies. Yet existing empirical evaluations of strategies to handle missing values have focused on inferential statistics. Results: Here we conduct a systematic benchmark of missing-values strategies in predictive models with a focus on large health databases: 4 electronic health record datasets, 1 population brain imaging database, 1 health survey, and 2 intensive care surveys. Using gradient-boosted trees, we compare native support for missing values with simple and state-of-the-art imputation prior to learning. We investigate prediction accuracy and computational time. For prediction after imputation, we find that adding an indicator to express which values have been imputed is important, suggesting that the data are missing not at random. Elaborate missing-values imputation can improve prediction compared to simple strategies but requires longer computational time on large data. Learning trees that model missing values—with missing incorporated attribute—leads to robust, fast, and well-performing predictive modeling. Conclusions: Native support for missing values in supervised machine learning predicts better than state-of-the-art imputation with much less computational cost. When using imputation, it is important to add indicator columns expressing which values have been imputed.
- 15 April 2022
- Published by Oxford University Press (OUP) in GigaScience
- Vol. 11
Top Cited PapersOpen AccessBenchmarking missing-values approaches for predictive models on health databases
Background: As databases grow larger, it becomes harder to fully control their collection, and they frequently come with missing values. These large databases are well suited to train machine learning models, e.g., for forecasting or to extract biomarkers in biomedical settings. Such predictive approaches can use discriminative—rather than generative—modeling and thus open the door to new missing-values strategies. Yet existing empirical evaluations of strategies to handle missing values have focused on inferential statistics. Results: Here we conduct a systematic benchmark of missing-values strategies in predictive models with a focus on large health databases: 4 electronic health record datasets, 1 population brain imaging database, 1 health survey, and 2 intensive care surveys. Using gradient-boosted trees, we compare native support for missing values with simple and state-of-the-art imputation prior to learning. We investigate prediction accuracy and computational time. For prediction after imputation, we find that adding an indicator to express which values have been imputed is important, suggesting that the data are missing not at random. Elaborate missing-values imputation can improve prediction compared to simple strategies but requires longer computational time on large data. Learning trees that model missing values—with missing incorporated attribute—leads to robust, fast, and well-performing predictive modeling. Conclusions: Native support for missing values in supervised machine learning predicts better than state-of-the-art imputation with much less computational cost. When using imputation, it is important to add indicator columns expressing which values have been imputed.
- Open Access
Benchmarking missing-values approaches for predictive models on health databases
BACKGROUND: As databases grow larger, it becomes harder to fully control their collection, and they frequently come with missing values: incomplete observations. These large databases are well suited to train machine-learning models, for instance for forecasting or to extract biomarkers in biomedical settings. Such predictive approaches can use discriminative -- rather than generative -- modeling, and thus open the door to new missing-values strategies. Yet existing empirical evaluations of strategies to handle missing values have focused on inferential statistics. RESULTS: Here we conduct a systematic benchmark of missing-values strategies in predictive models with a focus on large health databases: four electronic health record datasets, a population brain imaging one, a health survey and two intensive care ones. Using gradient-boosted trees, we compare native support for missing values with simple and state-of-the-art imputation prior to learning. We investigate prediction accuracy and computational time. For prediction after imputation, we find that adding an indicator to express which values have been imputed is important, suggesting that the data are missing not at random. Elaborate missing values imputation can improve prediction compared to simple strategies but requires longer computational time on large data. Learning trees that model missing values-with missing incorporated attribute-leads to robust, fast, and well-performing predictive modeling. CONCLUSIONS: Native support for missing values in supervised machine learning predicts better than state-of-the-art imputation with much less computational cost. When using imputation, it is important to add indicator columns expressing which values have been imputed.
Open AccessBenchmarking missing-values approaches for predictive models on health databases
BACKGROUND: As databases grow larger, it becomes harder to fully control their collection, and they frequently come with missing values: incomplete observations. These large databases are well suited to train machine-learning models, for instance for forecasting or to extract biomarkers in biomedical settings. Such predictive approaches can use discriminative -- rather than generative -- modeling, and thus open the door to new missing-values strategies. Yet existing empirical evaluations of strategies to handle missing values have focused on inferential statistics. RESULTS: Here we conduct a systematic benchmark of missing-values strategies in predictive models with a focus on large health databases: four electronic health record datasets, a population brain imaging one, a health survey and two intensive care ones. Using gradient-boosted trees, we compare native support for missing values with simple and state-of-the-art imputation prior to learning. We investigate prediction accuracy and computational time. For prediction after imputation, we find that adding an indicator to express which values have been imputed is important, suggesting that the data are missing not at random. Elaborate missing values imputation can improve prediction compared to simple strategies but requires longer computational time on large data. Learning trees that model missing values-with missing incorporated attribute-leads to robust, fast, and well-performing predictive modeling. CONCLUSIONS: Native support for missing values in supervised machine learning predicts better than state-of-the-art imputation with much less computational cost. When using imputation, it is important to add indicator columns expressing which values have been imputed.
- Open Access
- 1 January 2022
- Published by OpenEdition in Statistique et société
- Vol. 10 | 1, 19-36
ICUBAM (Intensive Care Unit Bed Availability Monitoring) Intensive care unit bed availability monitoring and analysis in the Grand Est region of France during the COVID-19 epidemic
Reliable information is an essential component of responding to a sudden and large disease outbreak such as COVID-19, particularly with respect to critical care beds (CCBs) availability. This article presents: i) the development and construction of ICUBAM, a tool that collects in real-time and visualizes information on CCB availability entered directly by intensivists; ii) an analysis and interpretation of the data collected over a 6-week period during the first wave of the epidemic in the hard-hit Grand Est region of France; iii) an analysis and interpretation of the data collected during the first wave of the epidemic in the Grand Est region; iv) the development of a medium and long term prediction using SEIR models, and a short term statistical model to predict the number of CCBs.Data ingested by ICUBAM were used to anticipate CCB shortages and predict future admissions. Most importantly, we demonstrate the importance of having a cross-functional team involving statisticians computer scientists and physicists working both with first-line medical responders and local health agencies and the importance of leveraging appropriate data. This allowed us to quickly implement effective tools to models the COVID-19 epidemic’s evolution and assist in critical decision-making processes.
- 1 January 2022
- Published by OpenEdition in Statistique et société
- Vol. 10 | 1, 19-36
Open AccessICUBAM (Intensive Care Unit Bed Availability Monitoring) Intensive care unit bed availability monitoring and analysis in the Grand Est region of France during the COVID-19 epidemic
Reliable information is an essential component of responding to a sudden and large disease outbreak such as COVID-19, particularly with respect to critical care beds (CCBs) availability. This article presents: i) the development and construction of ICUBAM, a tool that collects in real-time and visualizes information on CCB availability entered directly by intensivists; ii) an analysis and interpretation of the data collected over a 6-week period during the first wave of the epidemic in the hard-hit Grand Est region of France; iii) an analysis and interpretation of the data collected during the first wave of the epidemic in the Grand Est region; iv) the development of a medium and long term prediction using SEIR models, and a short term statistical model to predict the number of CCBs.Data ingested by ICUBAM were used to anticipate CCB shortages and predict future admissions. Most importantly, we demonstrate the importance of having a cross-functional team involving statisticians computer scientists and physicists working both with first-line medical responders and local health agencies and the importance of leveraging appropriate data. This allowed us to quickly implement effective tools to models the COVID-19 epidemic’s evolution and assist in critical decision-making processes.
- 1 April 2021
- Published by Elsevier in Anaesthesia Critical Care & Pain Medicine
- Vol. 40 (2) , 100827
Trauma reloaded: Trauma registry in the era of data science
- 1 April 2021
- Published by Elsevier in Anaesthesia Critical Care & Pain Medicine
- Vol. 40 (2) , 100827
Trauma reloaded: Trauma registry in the era of data science
- 16 July 2020
- Published by Springer Nature in Statistics and Computing
- Vol. 30 (6) , 1629-1643
Imputation and low-rank estimation with Missing Not At Random data
- 16 July 2020
- Published by Springer Nature in Statistics and Computing
- Vol. 30 (6) , 1629-1643
Imputation and low-rank estimation with Missing Not At Random data
- Open Access
Hydroxychloroquine with or without azithromycin and in-hospital mortality or discharge in patients hospitalized for COVID-19 infection: a cohort study of 4,642 in-patients in France
Objective: To assess the clinical effectiveness of oral hydroxychloroquine (HCQ) with or without azithromycin (AZI) in preventing death or leading to hospital discharge. Design: Retrospective cohort study. Setting: An analysis of data from electronic medical records and administrative claim data from the French Assistance Publique - Hôpitaux de Paris (AP-HP) data warehouse, in 39 public hospitals, Ile-de-France, France. Participants: All adult inpatients with at least one PCR-documented SARS-CoV-2 RNA from a nasopharyngeal sample between February 1 st , 2020 and April 6 th , 2020 were eligible for analysis. The study population was restricted to patients who did not receive COVID-19 treatments assessed in ongoing trials, including antivirals and immunosuppressive drugs. End of follow-up was defined as the date of death, discharge home, day 28 after admission, whichever occurred first, or administrative censoring on May 4, 2020. Intervention: Patients were further classified into 3 groups: (i) receiving HCQ alone, (ii) receiving HCQ together with AZI, and (iii) receiving neither HCQ nor AZI. Exposure to a HCQ/AZI combination was defined as a simultaneous prescription of the 2 treatments (more or less one day). Main outcome measures: The primary outcome was all-cause 28-day mortality as a time-to-event endpoint under a competing risks survival analysis framework. The secondary outcome was 28-day discharge home. Augmented inverse probability of treatment weighted (AIPTW) estimates of the average treatment effect (ATE) were computed to account for confounding. Results: A total of 4,642 patients (mean age: 66.1 ± 18; males: 2,738 (59%)) were included, of whom 623 (13.4%) received HCQ alone, 227 (5.9%) received HCQ plus AZI, and 3,792 (81.7%) neither drug. Patients receiving ‘HCQ alone’ or ‘HCQ plus AZI’ were more likely younger, males, current smokers and overall presented with slightly more co-morbidities (obesity, diabetes, any chronic pulmonary diseases, liver diseases), while no major difference was apparent in biological parameters. After accounting for confounding, no statistically significant difference was observed between the ‘HCQ’ and ‘Neither drug’ groups for 28-day mortality: AIPTW absolute difference in ATE was +1.24% (−5.63 to 8.12), ratio in ATE 1.05 (0.77 to 1.33). 28-day discharge rates were statistically significantly higher in the ‘HCQ’ group: AIPTW absolute difference in ATE (+11.1% [3.30 to 18.9]), ratio in ATE (1.25 [1.07 to 1.42]). As for the ‘HCQ+AZI’ vs neither drug, trends for significant differences and ratios in AIPTW ATE were found suggesting higher mortality rates in the former group (difference in ATE +9.83% [-0.51 to 20.17], ratio in ATE 1.40 [0.98 to 1.81];p=0.062). Conclusions: Using a large non-selected population of inpatients hospitalized for COVID-19 infection in 39 hospitals in France and robust methodological approaches, we found no evidence for efficacy of HCQ or HCQ combined with AZI on 28-day mortality. Our results suggested a possible excess risk of mortality associated with HCQ combined with AZI, but not with HCQ alone. Significantly higher rates of discharge home were observed in patients treated by HCQ, a novel finding warranting further confirmation in replicative studies. Altogether, our findings further support the need to complete currently undergoing randomized clinical trials. WHAT THIS PAPER ADDS?: What is already known on this subject: - The use of Hydroxychloroquine (HCQ) or HCQ with azithromycin (AZI) has been associated with viral load reduction at 6 days in COVID-19 infected patients - No difference between HCQ and no-HCQ groups in terms of risk of death or need for mechanical ventilation was found in two large cohorts of hospitalized COVID-19 infected patients What this study adds: - Using a large non-selected population of inpatients hospitalized for COVID-19 infection in 39 hospitals in France and robust methodological approaches, we found no evidence for efficacy of HCQ on 28-day mortality - Our results suggest an excess risk of mortality in patients treated by a combination of HCQ and AZI, but not with HCQ alone - Significantly higher rates of discharge home were observed in patients treated by HCQ, a novel finding warranting further confirmation in replicative studies
Open AccessHydroxychloroquine with or without azithromycin and in-hospital mortality or discharge in patients hospitalized for COVID-19 infection: a cohort study of 4,642 in-patients in France
Objective: To assess the clinical effectiveness of oral hydroxychloroquine (HCQ) with or without azithromycin (AZI) in preventing death or leading to hospital discharge. Design: Retrospective cohort study. Setting: An analysis of data from electronic medical records and administrative claim data from the French Assistance Publique - Hôpitaux de Paris (AP-HP) data warehouse, in 39 public hospitals, Ile-de-France, France. Participants: All adult inpatients with at least one PCR-documented SARS-CoV-2 RNA from a nasopharyngeal sample between February 1 st , 2020 and April 6 th , 2020 were eligible for analysis. The study population was restricted to patients who did not receive COVID-19 treatments assessed in ongoing trials, including antivirals and immunosuppressive drugs. End of follow-up was defined as the date of death, discharge home, day 28 after admission, whichever occurred first, or administrative censoring on May 4, 2020. Intervention: Patients were further classified into 3 groups: (i) receiving HCQ alone, (ii) receiving HCQ together with AZI, and (iii) receiving neither HCQ nor AZI. Exposure to a HCQ/AZI combination was defined as a simultaneous prescription of the 2 treatments (more or less one day). Main outcome measures: The primary outcome was all-cause 28-day mortality as a time-to-event endpoint under a competing risks survival analysis framework. The secondary outcome was 28-day discharge home. Augmented inverse probability of treatment weighted (AIPTW) estimates of the average treatment effect (ATE) were computed to account for confounding. Results: A total of 4,642 patients (mean age: 66.1 ± 18; males: 2,738 (59%)) were included, of whom 623 (13.4%) received HCQ alone, 227 (5.9%) received HCQ plus AZI, and 3,792 (81.7%) neither drug. Patients receiving ‘HCQ alone’ or ‘HCQ plus AZI’ were more likely younger, males, current smokers and overall presented with slightly more co-morbidities (obesity, diabetes, any chronic pulmonary diseases, liver diseases), while no major difference was apparent in biological parameters. After accounting for confounding, no statistically significant difference was observed between the ‘HCQ’ and ‘Neither drug’ groups for 28-day mortality: AIPTW absolute difference in ATE was +1.24% (−5.63 to 8.12), ratio in ATE 1.05 (0.77 to 1.33). 28-day discharge rates were statistically significantly higher in the ‘HCQ’ group: AIPTW absolute difference in ATE (+11.1% [3.30 to 18.9]), ratio in ATE (1.25 [1.07 to 1.42]). As for the ‘HCQ+AZI’ vs neither drug, trends for significant differences and ratios in AIPTW ATE were found suggesting higher mortality rates in the former group (difference in ATE +9.83% [-0.51 to 20.17], ratio in ATE 1.40 [0.98 to 1.81];p=0.062). Conclusions: Using a large non-selected population of inpatients hospitalized for COVID-19 infection in 39 hospitals in France and robust methodological approaches, we found no evidence for efficacy of HCQ or HCQ combined with AZI on 28-day mortality. Our results suggested a possible excess risk of mortality associated with HCQ combined with AZI, but not with HCQ alone. Significantly higher rates of discharge home were observed in patients treated by HCQ, a novel finding warranting further confirmation in replicative studies. Altogether, our findings further support the need to complete currently undergoing randomized clinical trials. WHAT THIS PAPER ADDS?: What is already known on this subject: - The use of Hydroxychloroquine (HCQ) or HCQ with azithromycin (AZI) has been associated with viral load reduction at 6 days in COVID-19 infected patients - No difference between HCQ and no-HCQ groups in terms of risk of death or need for mechanical ventilation was found in two large cohorts of hospitalized COVID-19 infected patients What this study adds: - Using a large non-selected population of inpatients hospitalized for COVID-19 infection in 39 hospitals in France and robust methodological approaches, we found no evidence for efficacy of HCQ on 28-day mortality - Our results suggest an excess risk of mortality in patients treated by a combination of HCQ and AZI, but not with HCQ alone - Significantly higher rates of discharge home were observed in patients treated by HCQ, a novel finding warranting further confirmation in replicative studies
- Open Access
Doubly robust treatment effect estimation with missing attributes
Missing attributes are ubiquitous in causal inference, as they are in most applied statistical work. In this paper, we consider various sets of assumptions under which causal inference is possible despite missing attributes and discuss corresponding approaches to average treatment effect estimation, including generalized propensity score methods and multiple imputation. Across an extensive simulation study, we show that no single method systematically out-performs others. We find, however, that doubly robust modifications of standard methods for average treatment effect estimation with missing data repeatedly perform better than their non-doubly robust baselines; for example, doubly robust generalized propensity score methods beat inverse-weighting with the generalized propensity score. This finding is reinforced in an analysis of an observations study on the effect on mortality of tranexamic acid administration among patients with traumatic brain injury in the context of critical care management. Here, doubly robust estimators recover confidence intervals that are consistent with evidence from randomized trials, whereas non-doubly robust estimators do not.
Open AccessDoubly robust treatment effect estimation with missing attributes
Missing attributes are ubiquitous in causal inference, as they are in most applied statistical work. In this paper, we consider various sets of assumptions under which causal inference is possible despite missing attributes and discuss corresponding approaches to average treatment effect estimation, including generalized propensity score methods and multiple imputation. Across an extensive simulation study, we show that no single method systematically out-performs others. We find, however, that doubly robust modifications of standard methods for average treatment effect estimation with missing data repeatedly perform better than their non-doubly robust baselines; for example, doubly robust generalized propensity score methods beat inverse-weighting with the generalized propensity score. This finding is reinforced in an analysis of an observations study on the effect on mortality of tranexamic acid administration among patients with traumatic brain injury in the context of critical care management. Here, doubly robust estimators recover confidence intervals that are consistent with evidence from randomized trials, whereas non-doubly robust estimators do not.
- Open Access
ICU Bed Availability Monitoring and analysis in the Grand Est region of France during the COVID-19 epidemic
Background: Reliable information is an essential component for responding to the COVID-19 epidemic, especially regarding the availability of critical care beds (CCBs). We propose three contributions: a) ICUBAM (ICU Bed Availability Monitor), a tool which both collects and visualizes information on CCB availability entered directly by intensivists. b) An analysis of CCB availability and ICU admissions and outcomes using collected by \ICUBAM during a 6-week period in the hard-hit Grand Est region of France, and c) Explanatory and predictive models adapted to CCB availability prediction, and fitted to availability information collected by ICUBAM. Methods: We interact directly with intensivists twice a day, by sending a SMS with a web link to the ICUBAM form where they enter 8 numbers: number of free and occupied CCBs (ventilator-equipped) for both COVID-19 positive and COVID-19- negative patients, the number of COVID-19 related ICU deaths and discharges, the number of ICU refusals, and the number of patients transferred to another region due to bed shortages. The collected data are described using univariate and multivariate methods such as correspondence analysis and then modeled at different scales: a medium and long term prediction using SEIR models, and a short term statistical model to predict the number of CCBs. Results: ICUBAM was brought online March 25, and is currently being used in the Grand-Est region by 109 intensivists representing 40 ICUs (95% of ICUs). ICUBAM allows for the calculation of CCB availability, admission and discharge statistics. Our analysis of data describes the evolution and extent of the COVID-19 health crisis in the Grand-Est region: on April 6th, at maximum bed capacity, 1056 ventilator-equipped CCBs were present, representing 211% of the nominal regional capacity of 501 beds. From March 19th to March 31st, average daily COVID-19 ICU inflow was 68 patients/day, and 314 critical care patients were transferred out of the Grand-Est region. With French lockdown starting on March 17th, a decrease of the daily inflow was found starting on April 1st: 23 patients/day during the first fortnight of April, and 7 patients/day during the last fortnight. However, treatment time for COVID-19 occupied CCBs is long: 15 days after the peak on March 31st, only 20% of ICU beds have been freed (50% after 1 month). Region-wide COVID-19 related in-ICU mortality is evaluated at 31%. Models trained from ICUBAM data are able to describe and predict the evolution of bed usage for the Grand-Estregion. Conclusion: We observe strong uptake of the ICUBAM tool, amongst both physicians and local healthcare stakeholders (health agencies, first responders etc.). We are able to leverage data collected with ICUBAM to better understand the evolution of the COVID-19 epidemic in the Grand Est region. We also present how data ingested by ICUBAM can be used to anticipate CCB shortages and predict future admissions. Most importantly, we demonstrate the importance of having a cross-functional team involving physicians, statisticians and computer scientists working both with first-line medical responders and local health agencies. This allowed us to quickly implement effective tools to assist in critical decision-making processes.
Open AccessICU Bed Availability Monitoring and analysis in the Grand Est region of France during the COVID-19 epidemic
Background: Reliable information is an essential component for responding to the COVID-19 epidemic, especially regarding the availability of critical care beds (CCBs). We propose three contributions: a) ICUBAM (ICU Bed Availability Monitor), a tool which both collects and visualizes information on CCB availability entered directly by intensivists. b) An analysis of CCB availability and ICU admissions and outcomes using collected by \ICUBAM during a 6-week period in the hard-hit Grand Est region of France, and c) Explanatory and predictive models adapted to CCB availability prediction, and fitted to availability information collected by ICUBAM. Methods: We interact directly with intensivists twice a day, by sending a SMS with a web link to the ICUBAM form where they enter 8 numbers: number of free and occupied CCBs (ventilator-equipped) for both COVID-19 positive and COVID-19- negative patients, the number of COVID-19 related ICU deaths and discharges, the number of ICU refusals, and the number of patients transferred to another region due to bed shortages. The collected data are described using univariate and multivariate methods such as correspondence analysis and then modeled at different scales: a medium and long term prediction using SEIR models, and a short term statistical model to predict the number of CCBs. Results: ICUBAM was brought online March 25, and is currently being used in the Grand-Est region by 109 intensivists representing 40 ICUs (95% of ICUs). ICUBAM allows for the calculation of CCB availability, admission and discharge statistics. Our analysis of data describes the evolution and extent of the COVID-19 health crisis in the Grand-Est region: on April 6th, at maximum bed capacity, 1056 ventilator-equipped CCBs were present, representing 211% of the nominal regional capacity of 501 beds. From March 19th to March 31st, average daily COVID-19 ICU inflow was 68 patients/day, and 314 critical care patients were transferred out of the Grand-Est region. With French lockdown starting on March 17th, a decrease of the daily inflow was found starting on April 1st: 23 patients/day during the first fortnight of April, and 7 patients/day during the last fortnight. However, treatment time for COVID-19 occupied CCBs is long: 15 days after the peak on March 31st, only 20% of ICU beds have been freed (50% after 1 month). Region-wide COVID-19 related in-ICU mortality is evaluated at 31%. Models trained from ICUBAM data are able to describe and predict the evolution of bed usage for the Grand-Estregion. Conclusion: We observe strong uptake of the ICUBAM tool, amongst both physicians and local healthcare stakeholders (health agencies, first responders etc.). We are able to leverage data collected with ICUBAM to better understand the evolution of the COVID-19 epidemic in the Grand Est region. We also present how data ingested by ICUBAM can be used to anticipate CCB shortages and predict future admissions. Most importantly, we demonstrate the importance of having a cross-functional team involving physicians, statisticians and computer scientists working both with first-line medical responders and local health agencies. This allowed us to quickly implement effective tools to assist in critical decision-making processes.