Clelia Colas’s research while affiliated with Capgemini and other places

Importance: Decision-making in trauma patients remains challenging and often results in deviation from guidelines. Machine-Learning (ML) enhanced decision-support could improve hemorrhage resuscitation. Aim: To develop a ML enhanced decision support tool to predict Need for Hemorrhage Resuscitation (NHR) (part I) and test the collection of the predictor variables in real time in a smartphone app (part II). Design, setting, and participants: Development of a ML model from a registry to predict NHR relying exclusively on prehospital predictors. Several models and imputation techniques were tested. Assess the feasibility to collect the predictors of the model in a customized smartphone app during prealert and generate a prediction in four level-1 trauma centers to compare the predictions to the gestalt of the trauma leader. Main outcomes and measures: Part 1: Model output was NHR defined by 1) at least one RBC transfusion in resuscitation, 2) transfusion ≥ 4 RBC within 6 h, 3) any hemorrhage control procedure within 6 h or 4) death from hemorrhage within 24 h. The performance metric was the F4-score and compared to reference scores (RED FLAG, ABC). In part 2, the model and clinician prediction were compared with Likelihood Ratios (LR). Results: From 36,325 eligible patients in the registry (Nov 2010-May 2022), 28,614 were included in the model development (Part 1). Median age was 36 [25-52], median ISS 13 [5-22], 3249/28614 (11%) corresponded to the definition of NHR. A XGBoost model with nine prehospital variables generated the best predictive performance for NHR according to the F4-score with a score of 0.76 [0.73-0.78]. Over a 3-month period (Aug-Oct 2022), 139 of 391 eligible patients were included in part II (38.5%), 22/139 with NHR. Clinician satisfaction was high, no workflow disruption observed and LRs comparable between the model and the clinicians. Conclusions and relevance: The ShockMatrix pilot study developed a simple ML-enhanced NHR prediction tool demonstrating a comparable performance to clinical reference scores and clinicians. Collecting the predictor variables in real-time on prealert was feasible and caused no workflow disruption.

Importance Decision-making in trauma patients remains challenging and often result in deviation from guidelines. Machine-Learning (ML) enhanced decision-support could improve hemorrhage resuscitation. Aim To develop a ML enhanced decision support tool to predict Need for Hemorrhage Resuscitation (NHR) (part I) and test the collection of the predictor variables in real time in a smartphone app (part II). Design, Setting, and Participants Development of a ML model from a registry to predict NHR relying exclusively on prehospital predictors. Several models and imputation techniques were tested. Assess the feasibility to collect the predictors of the model in a customized smartphone app during prealert and generate a prediction in four level-1 trauma centers to compare the predictions to the gestalt of the trauma leader. Main Outcomes and Measures Part 1: Model output was NHR defined by 1) at least one RBC transfusion in resuscitation, 2) transfusion ≥ 4 RBC within 6 hours, 3) any hemorrhage control procedure within 6 hours or 4) death from hemorrhage within 24 hours. The performance metric was the F4-score and compared to reference scores (RED FLAG, ABC). In part 2, the model and clinician prediction were compared with Likelihood Ratios (LR). Results From 36325 eligible patients in the registry (Nov 2010 - May 2022), 28614 were included in the model development (Part 1). Median age was 36 [25–52], median ISS 13 [5–22], 3249/28614 (11%) corresponded to the definition of NHR. A XGBoost model with nine prehospital variables generated the best predictive performance for NHR according to the F4-score with a score of 0.76 [0.73–0.78]. Over a 3-month period (Aug - Oct 2022), 139 of 391 eligible patients were included in part II (38.5%), 22/139 with NHR. Clinician satisfaction was high, no workflow disruption observed and LRs comparable between the model and the clinicians. Conclusions and Relevance The ShockMatrix pilot study developed a simple ML-enhanced NHR prediction tool demonstrating a comparable performance to clinical reference scores and clinicians. Collecting the predictor variables in real-time on prealert was feasible and caused no workflow disruption.

Background Emerging evidence suggests that the reallocation of health care resources during the COVID-19 pandemic negatively impacts health care system. This study describes the epidemiology and the outcome of major trauma patients admitted to centers in France during the first wave of the COVID-19 outbreak. Methods This retrospective observational study included all consecutive trauma patients aged 15 years and older admitted into 15 centers contributing to the TraumaBase® registry during the first wave of the SARS-CoV-2 pandemic in France. This COVID-19 trauma cohort was compared to historical cohorts (2017–2019). Results Over a 4 years-study period, 5762 patients were admitted between the first week of February and mid-June. This cohort was split between patients admitted during the first 2020 pandemic wave in France (pandemic period, 1314 patients) and those admitted during the corresponding period in the three previous years (2017–2019, 4448 patients). Trauma patient demographics changed substantially during the pandemic especially during the lockdown period, with an observed reduction in both the absolute numbers and proportion exposed to road traffic accidents and subsequently admitted to traumacenters (348 annually 2017–2019 [55.4% of trauma admissions] vs 143 [36.8%] in 2020 p < 0.005). The in-hospital observed mortality and predicted mortality during the pandemic period were not different compared to the non-pandemic years. Conclusions During this first wave of COVID-19 in France, and more specifically during lockdown there was a significant reduction of patients admitted to designated trauma centers. Despite the reallocation and reorganization of medical resources this reduction prevented the saturation of the trauma rescue chain and has allowed maintaining a high quality of care for trauma patients.

Background Evidence increases to suggest that the reallocation of health care resources during considerable the COVID-19 pandemic impacts considerably any health system. This study describes the epidemiology and the outcome of major trauma patients admitted to centers in France during the first wave of the COVID-19 outbreak. Methods This retrospective observational study included all consecutive trauma patients aged 15 years and older admitted into 15 centers participating to the TraumaBase ® registry in France during the first wave of the SARS-CoV-2 pandemic in France. ResultsOver a 4 years-study period, 5762 patients were admitted between the first week of February and mid-June. This cohort was split between patients admitted during the first 2020 pandemic wave in France (pandemic period, 1314 patients) and those admitted during the corresponding period in the three previous years (2017-2019, 4448 patients). Patient demographics changed substantially during the pandemic and more specifically during the lockdown period specially with a reduction in both absolute numbers admitted and the proportion of road traffic accidents (348 annually 2017-2019 [55.4 % of trauma admissions] vs 143 [36.8 %] in 2020 p<0.005). Mortality during the pandemic period and the difference between predicted and observed mortality was not different compared to the non-pandemic years. Conclusions During this first wave of COVID-19 in France, management of trauma patients admitted to regional Traumacenters was not significantly altered, despite medical resources being reallocated and reorganized. Mortality as well as prehospital and in hospital care remained stable throughout the period of the first pandemic wave despite a massive increase in demand for acute care beds.