Ver 20

Data entry, processing, and statistical analysis were carried out using MedCalc Ver. 20 (MedCalc). For descriptive statistics, mean and SD were used for continuous variables, while frequencies and percentages were utilized for categorical variables. Student's t‐test was used as a significance test for continuous variables, and the Chi‐square test was used for categorical variables. The receiver operating characteristic (ROC) curve analysis was used to evaluate the predictive value of thrombocytopenia for COVID‐19 outcomes. p Values less than 0.05 (5%) were considered statistically insignificant.

The minimum size of the sample was determined based on the formula:
where Nmin, minimum sample size; NPNP, population from which the sample is selected n = 2,131, α—level of confidence for the results = 95%, f—fraction size = 0.5 e—maximum error assumed 5% = 0.05 (Zalewska and Niemiro, 2022 (link)).
The statistical analysis of the material was conducted using MedCalc ver.20.104 package. Distributions of the variables were determined by means of Shapiro–Wilk test. The results were presented as average and standard deviation completed with the median. The analysis of differences was performed using nonparametric tests (Mann–Whitney U-test). To analyze the qualitative variables, Chi-squared test was used. Relationships between the variables were investigated using Spearman’s rank correlation tests.

Microsoft Office Excel 2016 (Los Angeles, CA, USA) in combination with XLSTAT (Version 2019) were used to process data. Statistical analysis was carried out using the STATISTICA 13 computer software (TIBCO Software, Inc., Palo Alto, CA, USA) and MedCalc ver. 20.104 software. Power analysis was conducted in G*Power to determine a sufficient sample size using an α value of 0.05, a power of 0.80, and a large effect size. The normality of variable distribution was checked with Shapiro–Wilk test. A descriptive analysis was carried out using the mean ± standard deviation. To evaluate the relationship between variables we used Pearson r correlation and linear regression equations. Levels of statistical significance were determined at p < 0.05. With a one-factor variance analysis (ANOVA), we tested the differences between variables and compared using Fisher’s Post Hoc Test. The risk of premature death (ABSI z-score) was calculated according to the methodology of Krakauer and Krakauer [20 (link)]. Premature mortality risk was classified into five categories: very low (<−0.868); low (between −0.868 and −0.272); average (between −0.272 and 0.229); high (between 0.229 and 0.798); and very high (>0.798) [20 (link)].