Background: Timely identification of deteriorating COVID-19 patients is needed to guide changes in clinical management and admission to intensive care units (ICUs). There is significant concern that widely used Early warning scores (EWSs) underestimate illness severity in COVID-19 patients and therefore, we developed an early warning model specifically for COVID-19 patients. Methods: We retrospectively collected electronic medical record data to extract predictors and used these to fit a random forest model. To simulate the situation in which the model would have been developed after the first and implemented during the second COVID-19 ‘wave’ in the Netherlands, we performed a temporal validation by splitting all included patients into groups admitted before and after August 1, 2020. Furthermore, we propose a method for dynamic model updating to retain model performance over time. We evaluated model discrimination and calibration, performed a decision curve analysis, and quantified the importance of predictors using SHapley Additive exPlanations values. Results: We included 3514 COVID-19 patient admissions from six Dutch hospitals between February 2020 and May 2021, and included a total of 18 predictors for model fitting. The model showed a higher discriminative performance in terms of partial area under the receiver operating characteristic curve (0.82 [0.80–0.84]) compared to the National early warning score (0.72 [0.69–0.74]) and the Modified early warning score (0.67 [0.65–0.69]), a greater net benefit over a range of clinically relevant model thresholds, and relatively good calibration (intercept = 0.03 [− 0.09 to 0.14], slope = 0.79 [0.73–0.86]). Conclusions: This study shows the potential benefit of moving from early warning models for the general inpatient population to models for specific patient groups. Further (independent) validation of the model is needed.

Purpose: Unclear recommendations in transfusion guidelines may possibly lead to inconsistency in treatment of patients admitted to the intensive care unit. This study aimed to uncover variation in red blood cell (RBC) transfusion decisions in the ICU worldwide. Methods: Members of the European Society of Intensive Care Medicine (ESICM) were requested to complete an online questionnaire which included four different hypothetical clinical scenarios. The scenarios represented patients with acute myocardial infarction (AMI), abdominal sepsis, traumatic brain injury (TBI) and post-surgical complications. Hemoglobin level was 7∙3 g/dL in all scenarios. The questionnaire explored the physicians' transfusion decision in each clinical scenario and identified patient characteristics that were most influential in the transfusion decision. Results: In total 211 members participated in the study, of whom 142 (67%) completed the entire survey. Most variation was observed in the clinical scenario of sepsis, in which 49% decided to transfuse and 51% decided not to. In the clinical scenarios of AMI, TBI and post-surgical complications this was respectively; 75/25%, 35/65% and 66/34%. Conclusions: Critical care physicians differed in outcome of RBC transfusion decisions and weighed patient characteristics differently. These findings indicate that variation in transfusion practice amongst critical care physicians exists.

Background and Objectives: Most guidelines recommend a restrictive transfusion trigger of 7 g/dl. It is unclear whether this resulted in more uniform transfusion practices. The primary objective was to uncover the extent of variation in transfusion decisions within four scenarios of critically ill patients among critical care physicians in the Netherlands. Materials and methods: An online survey comprising four different hypothetical clinical scenarios was sent to all members of the Dutch Society of Intensive Care. The scenarios represented patients with acute myocardial infarction (Hb 8·5 g/dl), abdominal sepsis (Hb 7·1 g/dl), traumatic brain injury (TBI) (Hb 7·9 g/dl) and post-surgical complications (Hb 7·3 g/dl). The questions explored the decision whether or not to transfuse and a ranking of clinical characteristics playing the most important role in the transfusion decision. Results: A total of 224 members (22%) participated in the study of whom 188 (84%) completed all questions. The percentages of respondents that decided to transfuse ranged from 25·9% in the scenario with TBI to 81·6% in the scenario with post-surgical complications. Most controversy was seen in the scenario with sepsis for which 43·2% decided to transfuse, whereas 56·8% decided not to. Haemoglobin level, diagnosis and haemodynamics were most important for the transfusion decision in all scenarios. Conclusions: Physicians decided differently on red-blood-cell transfusion given the clinical scenarios and weighed clinical characteristics differently in their transfusion decisions. These findings suggest there still is substantial variation in critical care transfusion practice.