CMStatistics 2022: Start Registration
View Submission - CMStatistics
B2005
Title: COVID-19 hospitalization and mortality after conditioning on osmolality and the effect of CA and Vitamin D Authors:  Ayse Ulgen - Nottingham Trent University (United Kingdom) [presenting]
Hakan Sivgin - Tokat Gaziosmanpasa University (Turkey)
Sirin Cetin - Amasya University (Turkey)
Meryem Cetin - Amasya University (Turkey)
Wentian Li - Robert S Boas Center for Genomics and Human Genetics (United States)
Abstract: Osmolality, the concentration of solute particles, can be estimated from the measurement of three other blood test variables (sodium, glucose, and urea), and was rarely used for a prognosis for COVID-19. As a result of the analysis of the data obtained from COVID-19 patients, we found osmolality to be an excellent prognostic biomarker for both mortality and hospitalization. On the other hand, both hypocalcemia and vitamin-D deficiency are also significantly associated with a higher mortality rate in our data, while only calcium and not vitamin D is associated with a higher hospitalization rate. Different types of tests, including logistic regression, t-test, Wilcoxon test, lead to the same conclusion. After conditioning on osmolality in multiple logistic regression, calcium level remains to be significantly associated with both mortality and hospitalization (p-value <0.001). However, vitamin D loses its association with mortality when conditioning on osmolality (significant only at 0.05 level).