Statistical package for the social sciences 25

Statistical Package for the Social Sciences 25.0 (IBM Inc.) was used for all statistical analyses. The Kolmogorov-Smirnov test was used to assess data normality, with normally and non-normally distributed quantitative variables being reported as mean ± SD and median (min-max), respectively, while categorical variables are given as n (%). The inter-observer agreement of quantitative data was evaluated via intraclass correlation coefficient (ICC). An ICC > 0.75 was considered to indicate a good agreement. Quantitative data were compared via independent sample t tests and Mann-Whitney U tests when normally and non-normally distributed, respectively. A P < .05 was the significance threshold. Receiver operating characteristic (ROC) curves were constructed for variables exhibiting significance in initial analyses, with AUC, specificity, sensitivity, and optimal cut-off values being calculated. Those parameters with P < .01 were then incorporated into logistic regression analyses to screen main factors for differentiating RO from ccRCC and chRCC and to establish predictive models. Sensitivity, specificity, and accuracy values were calculated for these models, with AUC (95% CI) being used to evaluate predictive performance.

All statistical analyses (sensitivity, specificity, negative predictive value, and positive predictive value) were performed on MedCalc Statistical Software version v19.4.1 (MedCalc Software, Ostend, Belgium) and Statistical Package for the Social Sciences 25.0 (Armonk, NY: IBM Corp.). Patients’ data were expressed as mean ± standard deviation for distributed data and percentage for categorical variables. Shapiro–Wilk test was used for the normally distributed continuous variables. To compare continuous measurements between the groups, one-way analysis of variance (with Bonferroni correction) was used for the normally distributed parameters and Kruskal–Wallis test (with Dunn’s posthoc test post hoc analysis) was used for those that were not normally distributed. To analyze the categorical variables, Pearson’s chi-square test was used if ≤20% of cells had expected values of <5. Fisher’s exact test or Monte Carlo exact test was used if >20% of cells had expected values of <5. P < .05 was considered to be statistically significant.

Data were analyzed using the program Statistical Package for the Social Sciences 25.0 (IBM, Armonk, NY: IBM Corp.). Continuous variables were expressed as mean ± standard deviation, median (interquartile range, IQR), and categorical variables as numbers (n) and percentages (%). When the parametric test assumptions were met, t-test was used to compare differences between independent groups. When parametric test assumptions were not met, Mann–Whitney U-test was used to compare differences between independent groups. The chi-squared or Fisher's exact probability tests were used to compare demographics. In all analyses, P <.05 was considered statistically significant.

All the data were incorporated in Statistical Package for the Social Sciences 25.0 (IBM Corporation, USA) for performing descriptive analysis. Experimental data were analyzed by one-way analysis of variance, appropriate and significant differences among different restaurants were made at 95% confidence level (p < 0.05) by Tukey’s multiple comparison test [22 ]. For each exposure scenario, Monte Carlo simulation was run using Microsoft excel@risk software version 7.5.0 (Palisade Corporation) sampling 10,000 iterations to measure the annual risk of infection.

Descriptive parameters such as means and standard deviations (SDs) for normally distributed continuous data and frequencies and percentages for categorical data were calculated. Nonnormally distributed data were expressed as medians and 25–75% interquartile ranges. The t-test or Mann–Whitney U-test was used to compare the difference of numerical data distribution between two groups. The Chi-square (χ²) test was used to compare differences of allergen distribution among groups (Fisher’s exact test was used when the expected count is less than 5), and the Spearman rank correlation analysis was used to evaluate the correlation between allergens. Histograms and Venn diagrams were used to show the distribution of positive rates and the cross-sensitization among multiple allergens. The binary logistic regression was used to analyze the risk factors of allergen sensitization in three kinds of animals, and forest map was used to show the risk ratio. The odds ratio (OR) value and 95% confidence interval (CI) were calculated. P values below 0.05 were considered statistically significant. All statistical analyses were performed using Excel 2019 (Microsoft^® Excel^® 2019), Statistical Package for the Social Sciences 25.0 (International Business Machines Corporation Corp., Armonk, NY), and GraphPad Prism 8 (GraphPad Software, inc.).