RESUMO
Objective: Delirium is commonly reported from the inpatients with Coronavirus disease 2019 (COVID-19) infection. As delirium is closely associated with adverse clinical outcomes, prediction and prevention of delirium is critical. We developed a machine learning (ML) model to predict delirium in hospitalized patients with COVID-19 and to identify modifiable factors to prevent delirium. Methods: The data set (n = 878) from four medical centers was constructed. Total of 78 predictors were included such as demographic characteristics, vital signs, laboratory results and medication, and the primary outcome was delirium occurrence during hospitalization. For analysis, the extreme gradient boosting (XGBoost) algorithm was applied, and the most influential factors were selected by recursive feature elimination. Among the indicators of performance for ML model, the area under the curve of the receiver operating characteristic (AUROC) curve was selected as the evaluation metric. Results: Regarding the performance of developed delirium prediction model, the accuracy, precision, recall, F1 score, and the AUROC were calculated (0.944, 0.581, 0.421, 0.485, 0.873, respectively). The influential factors of delirium in this model included were mechanical ventilation, medication (antipsychotics, sedatives, ambroxol, piperacillin/tazobactam, acetaminophen, ceftriaxone, and propacetamol), and sodium ion concentration (all p < 0.05). Conclusions: We developed and internally validated an ML model to predict delirium in COVID-19 inpatients. The model identified modifiable factors associated with the development of delirium and could be clinically useful for the prediction and prevention of delirium in COVID-19 inpatients.
RESUMO
OBJECTIVES: The purpose of this study was to evaluate the relationship of nonalcoholic fatty liver and cardiovascular risk factors. METHODS: This study was conducted to investigate the association of nonalcoholic fatty liver and cardiovascular risk factors for adult men (n=2976) and women (n=2442) who were over 19 years old, after excluding the HBsAg(+) or anti-HCV(+) patients and the men and women with increased alcohol intake (men: 40 g/week, women: 20 g/week). RESULTS: Compared with the normal liver subjects, the nonalcoholic fatty liver subjects showed a significantly increased frequency of abnormal systolic blood pressure (> or =120 mmHg), fasting blood sugar (> or =100 mg/dL), total cholesterol (> or =200 mg/dL), triglyceride (> or =150 mg/dL), high density lipoprotein cholesterol (<40 mg/dL), low density lipoprotein cholesterol (> or =130 g m/dL) and abdominal obesity in men, and all these measures were significantly increased in the women except for abnormal HDL cholesterol. After adjusting for the body mass index, age, smoking, exercise and a nonalcoholic liver, the odds ratios of an abnormal waist hip ratio were 1.35(95% Confidence Interval=1.05-4.72) in the mild fatty liver, 1.61(1.19-2.18) in the moderate fatty liver, 2.77(1.57-4.92) in the severe fatty liver compared with a normal liver. The adjusted odds ratios for abnormal fasting blood sugar were 1.26(1.03-1.53) in the mild fatty liver, 1.62(1.27-2.06) in the moderate fatty liver and 1.77(1.12-2.78) in the severe fatty liver. The adjusted odds ratios for abnormal triglyceride were 1.38(1.11-1.72) in the mild fatty liver, 1.73(0.33-2.24) in the moderate fatty liver and 1.91(1.17-3.10) in the severe fatty liver of men. Adjusted odds ratios for abnormal triglyceride were 1.50(1.04-2.15) in mild, 1.71(1.07-2.68) in moderate, 1.81(0.69-4.38) in severe fatty liver of women. CONCLUSIONS: The nonalcoholic fatty liver subjects had more cardiovascular risk factors compared with the normal liver subjects. Thus, prevention and treatment of the nonalcoholic fatty liver is necessary by lifestyle modifications such as restriction of alcohol intake, no smoking, exercise and adequate eating habits.
