Spss for macintosh

IBM SPSS for Macintosh (version 27.0, IBM Corp., Armonk, NY, USA) software package was used for statistical analyses. Categorical variables were cross-tabulated and analysed with Pearson’s Chi-squared test or Fisher’s exact test if expected values were below 5. Student’s t-test and Mann–Whitney U test were used to compare differences between study groups in continuous variables. Pairwise comparisons were performed as post hoc analyses for Pearson’s Chi-squared test using Z test and Dunn’s (1964) procedure for Kruskal–Wallis H test, both with a Bonferroni correction for multiple comparisons. Treatment in ICU, LOHS, and CRP level were separately selected as outcome variables for binomial logistic regression analysis; age, sex, current smoking, excess alcohol consumption or regular use of drugs, history of immunosuppressive disease or medication, or both, origin of infection in mandible, and SAG-positive microbial sample were selected as independent variables. LOHS and CRP level were dichotomized by group median, and age was categorized into tertiles. P values ˂ 0.05 were considered significant throughout the study.

Measurement data fitting normal distribution were expressed as mean ± standard deviation while data fitting non-normal distribution were expressed as median (interquartile range). T-test was adopted for normal distribution and rank sum test was used for non-normal distribution when comparing measurement data between the two groups. Chi-square test was used to compare the counting data between two groups and Fisher's exact probability method was used when the theoretical frequency was less than 1. The survival curve was calculated using the Kaplan-Meier method and evaluated with the log-rank test. Statistical significance was defined as P < 0.05 and all data were analyzed by SPSS for Macintosh (version 24.0; IBM, Armonk, NY).

The statistical analyses were made with IBM SPSS for Macintosh, version 19.0.0 (IBM Corp., Armonk, NY, USA). Mann–Whitney nonparametric U tests were used for comparing age and MMSE scores between the two groups, while a Pearson’s χ² test was used for comparing gender distribution and vascular risk factors. To adjust for the potentially confounding effects of age, continuous variables were log-transformed to obtain a normal distribution, before a general linear model analysis of covariance (ANCOVA) was performed for each biomarker, with age included as a co-variate in the analyses. We then performed ANCOVA analyses for each biomarker, with both age and body mass index (BMI) included as co-variates. Even though the gender distribution did not differ in a statistically significant way between the two groups, we also performed an ANCOVA analyses for each biomarker, with age, BMI and gender included as co-variates. Age, gender, and IGF and IGFBP levels were available in all cases, but BMI was only available in 47 controls and 88 patients with AD. Because of the high CVs for the analyses of IGF-I, we also excluded cases with a CV >20% in a separate analysis. Spearman’s correlation coefficient r_s was determined for bivariate correlation analyses.

The agreement between AI and Brixia scoring in relation to RT-PCR results and disease severity was assessed using the kappa statistic. Kappa values of 0 to 0.20, 0.21 to 0.40, 0.41 to 0.60, 0.61 to 0.80, and 0.81 to 1 were classified as slight, fair, moderate, substantial, and near-perfect agreement, respectively.
We adhered to the Standards for Reporting of Diagnostic Accuracy statement in our reporting of results. Statistical analyses were performed using SPSS for Macintosh (version 27; IBM Corp), MedCalc for Windows (version 20.114; MedCalc Software Ltd), and Stata Statistical Software for Macintosh (version 14; StataCorp LP)