Sas enterprise guide 6.1 m1

A summary of demographic and baseline characteristics was constructed using descriptive analysis; the mean, maximum, minimum, and SD for quantitative variables, and the frequency and percentage for qualitative variables. Prevalence of IA with respect to the status of AD was analyzed using χ² tests or Fisher’s exact test. The logistic regression analyses were performed for IA and confounders (sex, age groups, hypertension, and diabetes mellitus) to calculate the odds ratio (OR) and 95% CI without considering any interactions. One of the co‐authors, a medical statistician, was tasked with supervision of the overall analytics procedure. All statistical analyses were performed using SAS Enterprise Guide 6.1 M1 (SAS Institute Inc., Cary, NC) and SPSS software package for Windows version 19.0 (IBM, Armonk, NY). All tests were 2‐sided, and P values <0.05 were deemed statistically significant.

Despite being based on administrative data sources, we calculated sample sizes to ensure that the study analysis was feasible. Based on a pilot study, to detect a difference in the incidence of PHN between the groups, the estimated sample size with a type I error of 0.05 and a power of 90% was 246.
The study endpoint was the date of PHN diagnosis, end of follow-up (180 days from the index date), or death. Log-rank tests were used to assess differences in the incidence of PHN, while Kaplan–Meier curves were used to calculate the 180-day PHN incidence rates between the cohort groups. After adjusting for disease-related potential confounders (respiratory disease, DM, cancer, autoimmune disease, and severe immunosuppressive status), stratified Cox proportional hazards modeling (stratified by sex, age, residence urbanization level, household income level, disability, number of outpatient visits, hospitalization) was used to calculate hazard ratios (HR) and corresponding 95% confidence intervals (CI) for subsequent PHN. The HRs associated with potential confounding factors were confirmed using a stratified Cox proportional hazards model. SAS Enterprise Guide 6.1M1 (SAS Institute Inc., Cary, NC, USA) was used for all statistical analyses. Statistical significance was set at p<0.05.