Sas software package 9

In order to highlight the variability observed between the 36 studied extracts by HPLC-DAD analyses, the peak areas of main phenolic compounds were then submitted to an exploratory chemometric analysis. To achieve it, a principal component analysis was applied on the dataset made of 36 samples × 8 variables (major extract constituents) [20 ]. The analysis was performed with SAS software package 9.4 (SAS Institute Inc., Cary, NC, USA).

All analyses were performed with SAS software package 9.4 (SAS Institute, Cary, NC). Owing to the extensive nature of the objectives and number of analyses, descriptive statistics were applied. Summary measures such as means and standard deviations (SDs) for continuous variables and counts, frequencies, and percentages for categorical variables were reported; no statistical significance tests were applied. The correspondence of patient- and HCP-reported data was evaluated for all similar outcomes from these 2 databases, including OIC symptoms and treatment use. All data were analyzed without imputation for missing responses.

Statistical analyses were performed using the SAS Software package 9.4 (SAS Institute Inc., Cary, NC, USA). Quantities of trichome secretion components were analyzed by analysis of variance (ANOVA), and the means were compared using Duncan’s multiple range test. Exit ratios for each concentration of each fraction evaluated in the bridge bioassay were tested for homogeneity between the two bridges in the arena by χ². In all cases, exit ratios were homogeneous between two bridges. EC₅₀ and EC₉₀ values were then estimated by use of PROC PROBIT, (SAS Institute 9.4 version), as described by Snyder et al. (2011) (link). EC is an abbreviation for Effective Concentration, and an EC₅₀ value is the predicted concentration at which 50% of the tested individuals respond to a stimulus; an EC₉₀ value is the predicted concentration at which 90% of tested individuals respond.

Descriptive statistics were used to summarize data. The annual number of reports of serious ADEs was determined from 2006 through 2014. Data from a previous study were used to determine the trend of reports of serious ADEs before 2006.^{4 (link)} Frequencies were determined according to each outcome category or drug. The proportion of ADE reports by age group was estimated based on the total number of serious ADE reports and compared against the proportion of drug use attributable to each age group.⁷ All analyses were conducted using SAS software package 9.4 (SAS Institute, Cary, NC).

The PROC GLM procedure in the SAS software package 9.4 (SAS Institute Inc.) was used for Analysis of Variance (ANOVA) analysis. Broad sense heritability within and across years was estimated from the ANOVA table using the formulas h² = σ _g²/(σ _g²+ σ _E²/r) and h² = σ _g²/(σ _g²+ σ_gxy²/y+σ _E²/ry), respectively, with σ _g² = genetic variance, σ_gxy² = genotype-by-environment interaction variance, σ _E² = error variance, r = number of replicates and y = number of years. The LSMEANS function in PROC MIXED was used to calculate the mean NB severity, mean DH and mean PH of each line. To determine whether DH and PH influence the disease development under field conditions, the mean NB severity of every line in every year was regressed to the mean DH and mean PH in the corresponding year using the PROC REG procedure. PH was found to have a significant impact in 2014 and both scorings in 2015 and was used as a covariate in QTL mapping. The Pearson correlation coefficients were calculated with the PROC CORR function.

Descriptive statistics were used to assess differences in demographics and beverage consumption between students who engage in smoking behaviours and students who do not (Student’s t-test for continuous variables and chi-square test for categorical variables). If variables were skewed, the Wilcoxon rank-sum test was used. All analyses were run using SAS software package 9.4 (Cary, NC).
Four separate models were developed for 1) current/former smokers; 2) current e-cigarette users; 3) days smoked cigarettes per month, and 4) days used e-cigarette per month. All models were adjusted for age, sex, BMI, school median income, ethnicity and school clustering. Models 1 and 2 used multiple logistic regression and model 3 and 4 used ordinal logistic regression to investigate the ability of beverage consumption to explain the variability in smoking behaviour. The significance level was set at 0.05 (two-sided test). Each beverage type was tested individually in models. For all beverage variables, 0 was the referent. In smoking models, the non-smoker group was the referent. In e-cigarette models, the non-e-cigarette user group was the referent.

Patient characteristics were compared across dispensing channels for all 3 therapeutic classes using chi-square tests for categorical variables and analysis of variance for continuous variables. We also described the proportion of patients on specialty drugs by therapeutic class and type of dispensing channel (see Figure 2). In the unadjusted analyses, multiple comparison tests were used to compare weighted adherence rates across channels for all patients and by therapeutic class. Multivariable (binary) logistic regression was used to assess the association between dispensing channel and adherence to specialty drugs after controlling for other covariates. The OOP cost and LIS status were highly correlated, so only the OOP cost was included in the multivariable analyses. All analyses were conducted at an α level of 0.05 using SAS software package 9.4 (SAS Institute, Cary, NC) and Microsoft Excel 2016 (Redmond, WA).