Evaluation of Risk Factors for Predicting Mortality in Critically Ill Adult COVID-19 Patients: A Retrospective Cohort Study

Abstract

Background and Objective: Various risk factors have been evaluated to predict the mortality associated with COVID-19. We aim to explore and compare the clinical and laboratory risk factors with various outcomes of the disease between survivors and non-survivors amongst patients with moderate to severe COVID-19 disease. Methods: All COVID-19 adult (≥ 18 years old) ICU in-patients with a definite outcome i.e. either death or discharge were included. The demographic, clinical, laboratory, treatment, and outcome data were retrieved. To explore the association between factors, univariate and multivariate logistic regression were done. Results: A total of 163 patients were included out of which eight patients were shifted to other hospitals. Finally, a total of 145 patients were included in the study. Out of 145, 47 patients didn’t survive and 98 survived. A significant proportion (85%) of non-survivors were current smokers and were observed to have COPD, HT, and CAD as comorbidities when compared to the survivors. The qSOFA score, CRP, and TLC counts were included in the multivariable logistic regression with the former being the independent risk factor. Interpretation and Conclusion: Our study highlighted that older age, higher SOFA score, increased levels of total leucocyte count, and C-Reactive protein were independent risk factors affecting the clinical outcome in moderate to severe COVID-19 patients; however, qSOFA was found to be an independent risk factor predictor.

Racial/Ethnic Differences in Chronic Disease Predictors Among American High School Students*

“…17 Each class was repeated 50 times starting with random starting values generated by mplus version 7.4. 2 This is an evidence-based method used to include a close examination of model fit indices for the estimated latent classes. 2 The final number of classes were determined based on conceptual meaning, the smallest estimated class proportions, entropy, and best fit criteria (Akaike Information Criterion [AIC], Bayesian Information Criterion [BIC], and adjusted BIC).…”

“…2 This is an evidence-based method used to include a close examination of model fit indices for the estimated latent classes. 2 The final number of classes were determined based on conceptual meaning, the smallest estimated class proportions, entropy, and best fit criteria (Akaike Information Criterion [AIC], Bayesian Information Criterion [BIC], and adjusted BIC). 17 Mplus produces maximum likelihood estimates to account for missing data and to estimate parameters.…”

“…Mplus also accounts for complex survey design by adjusting standard errors and the chi-square model fit test. 2 Bivariate analysis and multivariate logistic regression. After identifying the appropriate number of latent classes, cross-tabulations (bivariate analysis through chi-square tests) were performed to assess associations with 2 key predictors of chronic diseases (sleep and physical activity) and the main demographic characteristics of interest (sex and race).…”

“…28 Black males and females were more likely to sleep 5 or fewer hours on school nights than their white counterparts. 2 Cognitive development, performing health behaviors, and preventing chronic disease can be enhanced with the achievement of optimal sleep levels in adolescents. Consistent with previous research, the results of this study found Asians and blacks had significantly high odds of being classified as high risk for chronic disease development compared to whites.…”

“…Since 2020, COVID‐19 has been a leading cause of mortality worldwide. In addition, the presence of chronic diseases has been named as a prominent risk factor for hospitalization or death related to COVID‐19 2 …”