Sas 9.4 university edition

Descriptive and analytical statistics were generated by SAS 9.4 University Edition (SAS Institute, Cary. North Carolina, USA). The comparison of means was done using the paired sample t-tests. ANOVA was used to assess the mean difference of mJOA score at serial followups. Pearson's correlation coefficient was utilized to assess colinearity.

A sample size of 39 per group size was calculated based on a 90% power to assess for the lower limit of a one-sided 95% confidence interval above the non-inferiority limit of 10%, based on a standard deviation of total CPR score of 15%. This standard deviation was based on pilot data obtained prior to trial commencement and was not included in the study. In order to account for a higher standard deviation, we inflated our sample size to account for a standard deviation as high as 19%, resulting in a conservative sample size of 62 per group. The non-inferiority margin is difficult to define as depending on the resource setting a higher non inferior margin may be acceptable if no alternative trainer is available. Therefore we chose a non-inferiority margin that was based on similar studies assessing non-inferiority of different training methodologies for CPR.8 (link), 9 (link)
We calculated the absolute frequencies for categorical variables and the mean and standard deviation (SD) for the continuous variables. Categorical variables were analysed using a Chi-square and continuous variables using a T-test. Statistical analysis was performed using SAS 9.4 University edition.

All study measures were summarized using descriptive statistics. Medication utilization and insurance characteristics were described using the data collected on the last refill (most recent). Descriptive statistics were calculated for PDC for the overall patient cohort, adults, children, patients on primary insurance, and patients on both primary and secondary insurances. PDC means were stratified based on the three CFTR modulators. PDC means were also compared for children/adolescents and adults, patients with primary insurance, and those with both primary and secondary insurances. Independent t-test for continuous variables and chi-square for categorial variables were conducted and an a priori p < 0.05 was considered statistically significant. All analyses were performed using the SAS 9.4 University Edition (SAS Institute, Cary, NC).