Sas stat

All statistical analysis was performed using SAS/STAT (SAS Institute Inc., Cary,
North Carolina, USA).
Continuous variables with normal distribution were presented as mean and standard
deviation, and continuous variables with non-normal distribution were presented
as medians and interquartile ranges. Categorical variables were presented as
proportions. The Shapiro-Wilk test was performed as a normality test.
Multivariate logistic regression was used to estimate associations between
carotid atherosclerosis and clinical variables.
Multiple linear regression was utilized to analyze associations between arterial
stiffness and clinical variables or the presence of carotid atherosclerosis.
The Wilcoxon-Mann-Whitney test was used to compare two groups of non-parametric
results. The unpaired Student t test was applied for parametric
results.
One-way ANOVA was employed to compare the groups in FCRS classification.
All tests were two-tailed, and significance was set at p < 0.05.

Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated based on true positives, true negatives, false positives and false negatives as stated by [20 ] comparing results from on-farm culture of milk samples collected between April and July 2014 and reference laboratory results using the same milk samples. In addition, accuracy was calculated by dividing the number of true positives and true negatives by the total number of tests. The simple Cohen's kappa coefficient (κ) was calculated using the FREQ procedure of SAS version 9.3 (SAS/STAT, SAS Institute Inc., Cary, NC). This parameter assumes that the two response variables (on-farm culture system and gold standard) are independent ratings, and the coefficient equals 1 when there is complete agreement between the two tests. The null hypothesis for this test is that if agreement happens due to chance the Kappa coefficient is equal to zero. Under this null hypothesis, P-values associated with this test equal or smaller than 0.05 were considered significant.

The primary aim of the research described here was to compare PK profiles of metformin under fasting and fed conditions when administered as metformin eicosapentaenoate versus separately as metformin and icosapent ethyl. The profiles were compared via non-compartmental PK parameters as area under the curve AUC_last and C_max. The analysis consisted of fitting two linear mixed effects models, each using one of the primary outcomes as the response variable. The models both contained fixed effects for treatment group, study condition (fasting/fed), and the interaction of these two variables. In addition, a random participant effect was included to account for within-subject correlation between repeated measurements taken under the fasting and fed conditions. Least squares means (LSM) were obtained from the models in order to compare the effects of the drug treatment groups under each study condition and to compare the effects of the study conditions within each treatment group. LSM were compared using two-sample t-tests. Since this study was exploratory in nature, two-sided t-tests were performed for each pair-wise comparison. All analyses were carried out using SAS/STAT^® software, Version 9.4 of the SAS System for Windows (Cary, NC, USA), and all tests were evaluated using significance level α = 0.05. A result was considered statistically significant if p < α.

Paired t-tests (with Bonferroni corrected acceptance levels) were used to determine if the relative abundance of polymers described from environmental samples were significantly different to that expected from relative global production [2 ], waste production [9 (link)] or from polymer-specific modelled release of microplastics into surface waters [3 (link)]. To ensure normality, all data were arcsine transformed before use. MANOVA (multivariate analysis of variance) was used in SAS/STAT^® (SAS Institute Inc., Cary, North Carolina, USA) to determine if the relative abundance of polymers described differed among the three matrices, water, sediment and biota, with Tukey’s post hoc tests being used to identify where significant differences lay within polymers. Again, all data were arcsine transformed before use.

Normality was assessed with probability plots. Concentrations of biomarkers were compared between positive and negative conditions (i.e., with and without IL-1β, respectively) within each combination of culture type group (OCS and cartilage) and time point (48 versus 96 h) using Friedman’s chi-square with horse as a blocking factor (SAS/STAT, SAS Institute, Cary, NC, USA). A logarithmic (base e) transformation was applied to the fold changes before any downstream analyses. Effects of culture type and time on the log fold changes were assessed using mixed model analysis of variance. Where appropriate P-values were adjusted for multiple comparisons using Bonferroni’s procedure. The linear model specified culture group, time, and interaction between group and time as fixed effects. Denominator degrees of freedom for the fixed effects were approximated using the Kenward–Roger method. Horse identification was specified as the random effect. Within the specified interaction, the following comparisons were extracted: (1) time point 48 versus time point 96 for each group and (2) OCS versus cartilage at each time point. For all analysis of variance models, residuals were inspected to verify that the errors followed a normal distribution with constant variance. Values of P < 0.05 were considered significant.