ABSTRACT
Pulmonary fibrosis is one of the most severe long-term consequences of COVID-19. Corticosteroid treatment increases the chances of recovery; unfortunately, it can also have side effects. Therefore, we aimed to develop prediction models for a personalized selection of patients benefiting from corticotherapy. The experiment utilized various algorithms, including Logistic Regression, k-NN, Decision Tree, XGBoost, Random Forest, SVM, MLP, AdaBoost, and LGBM. In addition easily human-interpretable model is presented. All algorithms were trained on a dataset consisting of a total of 281 patients. Every patient conducted an examination at the start and three months after the post-COVID treatment. The examination comprised a physical examination, blood tests, functional lung tests, and an assessment of health state based on X-ray and HRCT. The Decision tree algorithm achieved balanced accuracy (BA) of 73.52%, ROC-AUC of 74.69%, and 71.70% F1 score. Other algorithms achieving high accuracy included Random Forest (BA 70.00%, ROC-AUC 70.62%, 67.92% F1 score) and AdaBoost (BA 70.37%, ROC-AUC 63.58%, 70.18% F1 score). The experiments prove that information obtained during the initiation of the post-COVID-19 treatment can be used to predict whether the patient will benefit from corticotherapy. The presented predictive models can be used by clinicians to make personalized treatment decisions.
ABSTRACT
BACKGROUND: The practical experiences of active pharmacists involved in managing critically ill patients with coronavirus disease 2019 (COVID-19) have been rarely reported. OBJECTIVE: This work aimed to share professional experiences on medication optimization and provide a feasible reference for the pharmaceutical care of critically ill patients with COVID-19. METHODS: This study was conducted in a COVID-19-designated hospital in China. A group of dedicated clinical pharmacists participated in multidisciplinary rounds to optimize the treatments for critically ill patients with COVID-19. Consensus on medication recommendations was reached by a multidisciplinary team through bi-daily discussion. Related drug, classification, cause, and adjustment content for recommendations were recorded and reviewed. RESULTS: A total of 111 medication recommendations were supplied for 22 out of 33 (56.7%) critically ill patients from 1 February 2020 to 18 March 2020, and 106 (95.5%) of these were accepted. Among these recommendations, 64 (67.7%), 32 (28.8%), and 15 (13.5%) were related to antibiotics and antifungals, antiviral agents, and other drugs, respectively. Recommendation types significantly differed for different anti-infectives (p < 0.05). For antibiotics and antifungals, treatment effectiveness accounted for 60.9% of recommendation types, with 15 (38.5%) cases related to untreated infections. For antiviral agents, adverse drug events were the most common recommendation types (84.4%), with 20 (74.1%) cases related to liver function dysfunction. Discontinuation of suspected antiviral agents (66.7%) was usually recommended after the occurrence of adverse events that may progress and bring poor outcomes. CONCLUSION: Forceful and extensive on-ward participation is recommended for clinical pharmacists in managing critically ill patients. Our experiences highlight the need for special attention toward untreated infections and adverse events related to antiviral agents.