Sas stat system for windows version 9

Descriptive statistics were used to characterize the patients, by level of AD severity, using numbers and percentages for categorical variables, and means and standard deviations (SDs) for continuous variables. Differences in baseline demographics among groups were assessed by χ² tests, t-tests, or one-way analysis of variance (ANOVA). Crude values for the 4 types of impairment scores were presented, by level of AD severity, as numbers and percentages for the patients with 0 scores, and as means, SDs, medians, and interquartile ranges (IQRs) for the patients with non-zero scores. Regression-adjusted values of the impairment scores were computed using 2-part generalized linear models with beta distributions with the level of AD severity as the main independent variable and while controlling for sex, age, body mass index (BMI), education, marital status, and household income. Differences in the mean adjusted values of the impairment scores among the different levels of AD severity were then assessed by ANOVA. A statistical significance of 0.05 was adopted.
All analyses were performed using SAS/Stat system for Windows, version 9.4 (SAS Institute, Cary, NC, USA).

Descriptive statistics for study participants who completed patient interviews were reported previously by Chan et al. (7 (link)). Crude values for each OOP category are presented according to disease severity, based on the unadjusted proportions of costs incurred and unadjusted OOP costs for those with positive values. To account for the zeroness and right-skewing nature of the distribution of costs, OOP costs were analysed using a 2-part model consisting of logistic regression to predict the probability of costs incurred and a generalized linear model to predict costs among those with positive values. Covariates were entered into the models to predict OOP costs, including sex, age, body mass index, education, marital status, and household income. Differences in mean adjusted values for each OOP cost category among severity levels were assessed by analysis of variance. All costs were reported in 2018 New Taiwan dollars (NT $30.58 = US$1).
Findings with a 2-sided p-value of ≤ 0.05 were considered significantly different between groups. All analyses were performed using SAS/Stat system for Windows, version 9.4 (SAS Institute, Cary, NC, USA).