Machine Learning Identifies ICU Outcome Predictors in a Multicenter COVID-19 Cohort. (preprint)

Background. Intensive Care Resources are heavily utilized during the COVID-19 pandemic. However, risk stratification and prediction of SARS-CoV-2 patient clinical outcomes upon ICU admission remain inadequate. This study aimed to develop a machine learning model, based on retrospective & prospective clinical data, to stratify patient risk and predict ICU survival and outcomes.Methods. A Germany-wide electronic registry was established to pseudonymously collect admission, therapeutic and discharge information of SARS-CoV-2 ICU patients retrospectively and prospectively. Machine learning approaches were evaluated for the accuracy and interpretability of predictions. The Explainable Boosting Machine approach was selected as the most suitable method. Individual, non-linear shape functions for predictive parameters and parameter interactions are reported. Results. 1,039 patients were included in the Explainable Boosting Machine model, 596 patients retrospectively collected, and 443 patients prospectively collected. The model for prediction of general ICU outcome was shown to be more reliable to predict “survival”. Age, inflammatory and thrombotic activity, and severity of ARDS at ICU admission were shown to be predictive of ICU survival. Patients’ age, pulmonary dysfunction and transfer from an external institution were predictors for ECMO therapy. The interaction of patient age with D-dimer levels on admission and creatinine levels with SOFA score without GCS were predictors for renal replacement therapy. Conclusions. Using Explainable Boosting Machine analysis, we confirmed and weighed previously reported and identified novel predictors for outcome in critically ill COVID-19 patients. Using this strategy, predictive modeling of COVID-19 ICU patient outcomes can be performed overcoming the limitations of linear regression models.Trial registration. “ClinicalTrials” (clinicaltrials.gov) under NCT04455451

COVID-19 vaccination acceptance among healthcare workers in Germany (preprint)

BackgroundVaccination hesitancy is a serious threat to achieve herd immunity in a global and rapidly changing pandemic situation. Health care workers play a key role in the treatment of patients with Coronavirus disease 2019 (COVID-19) and in promoting vaccination in the general population. The aim of the study was to provide data on COVID-19 vaccination acceptance and barriers among healthcare workers in Germany to support health policymakers choosing specific vaccination campaign strategies. MethodsAn online survey was conducted among health care workers in Germany in February 2021. The survey included 55 items on demographics, previous vaccination behavior, trust in vaccines, physicians, pharma industry, and health politics as well as fear of adverse effects, assumptions on disease consequences, knowledge about vaccines, information seeking behavior and a short COVID-19 vaccine knowledge test. ResultsA total of 4500 surveys could be analyzed. The overall vaccination acceptance was 91.7%. The age group [≤]20 years showed the lowest vaccination acceptance of all age groups. Regarding professional groups, residents showed the highest vaccination acceptance. Main factors for vaccination hesitancy were lack of trust in authorities and pharmaceutical companies. Personal and professional environment influenced the attitude towards a vaccination too. Participants with vaccination hesitancy were more likely to obtain information about COVID-19 vaccines via messenger services or online video platforms and underperformed in the knowledge test. ConclusionsIn conclusion, we found a high acceptance rate amongst German health care workers. Furthermore, several factors associated with vaccination hesitancy were identified which could be targeted in vaccination campaigns.

Staff Shortage in German Intensive Care Units During the COVID-19 Pandemic - Not only a Sensed Dilemma: Results from a Nationwide Survey (preprint)

Background: The surge in patients during the COVID-19 pandemic has exacerbated the looming problem of staff shortage in German ICUs possibly leading to worse outcomes for patients. Methods: Within the German Evidence Ecosystem CEOsys network, we conducted an online national mixed-methods survey assessing the standard of care in German ICUs treating patients with COVID-19. Results: A total of 171 German ICUs reported a median ideal number of patients per intensivist of 8 (interquartile range, IQR = 3rd quartile - 1st quartile = 4.0) and per nurse of 2.0 (IQR = 1.0). For COVID-19 patients, the median target was a maximum of 6.0 (IQR = 2.0) patients per intensivist or 2.0 (IQR = 0.0) patients per nurse. Targets for intensivists were rarely met by 15.2% and never met by 3.5% of responding institutions. Targets for nursing staffing could rarely be met in 32.2% and never in 5.3% of responding institutions. Conclusions: Shortages of staffing in the critical care setting are eminent during the COVID-19 pandemic and might not only negatively affect patient outcomes, but also staff wellbeing and healthcare costs. A joint effort that scrutinizes the demands and structures of our health care system seems fundamental to be prepared for the future.