Stata intercooled version 12

Both clinical, laboratory and radiological data were analyzed using STATA intercooled version 12.1. Quantitative data were analyzed, using student t-test to compare means of two groups. When the data was not normally distributed Mann-Whitney test was used. Qualitative data were compared using Chi square test. P value was considered significant if it was less than 0.05.

Data was analyzed by STATA intercooled version 12.1 using descriptive statistics and Independent-samples t-test, Mann-Whitney U test, Chi square, and Fisher’s exact test. Graphs were produced by using Excel or STATA program. P value was considered significant if it was less than 0.05.

Stata Intercooled, Version 12 (Stata Corporation, College Station, TX) was used for statistical analysis. Descriptive statistics comparing patients with and without 30-day readmission was analyzed by 2-sided Student’s t tests and χ² tests for continuous and categorical variables, respectively. When necessary, the Fischer exact test and Wilcoxon-Mann-Whitney rank sum test were used for small cell counts with fewer than 10 observations. Multivariable parametric binary logistic regression models were constructed to test the association of combined predictors with readmission within 30 days. Risk factors from the univariate analysis with P < .1 and cell counts greater than 20 observations each were included into the multivariate regression model. Variables that would result in absolute readmission (ie, reoperation, sepsis, pulmonary embolism, etc) were excluded from final multivariable analysis. A forward and backward stepwise regression analysis was conducted to identify risk factors associated with 30-day readmission. Odds ratios (ORs) are presented with 95% confidence intervals (CIs). The likelihood ratio (LR) test was implemented to compare the goodness of fit between final logistic regression models.