Stata special edition v 15.0 for windows

Demographic and baseline characteristics were evaluated using descriptive statistics. Continuous variables are reported as mean±SD or median (IQR) as appropriate for their distribution as determined by Shapiro-Wilk test. Categorical variables are presented as percentages. The descriptive characteristics of the three tertiles were compared using one-way analysis of variance for continuous variables, and Pearson’s χ² test for categorical variables. The study outcomes were calculated as end of IMWM program minus baseline values. A non-parametric test for trend across ordered tertiles, an extension of the Wilcoxon rank-sum test, was used to explore the relationship between the tertile of SMBG frequency and change in metabolic outcomes. Univariate and multivariable linear regressions were used to investigate the association between SMBG frequency and changes in body weight and A1C. A two-sided p value <0.05 was considered statistically significant. Statistical analysis was conducted using STATA Special Edition V.15.0 for Windows (StataCorp, College Station, Texas, USA, 2017).

This study was designed on an intent-to-treat principle and all available data were used for this analysis. There was no evidence that missing data were dependent on the study group. An unstructured covariance matrix was used to account for correlation between repeated outcomes. Χ² test, Fisher’s exact test and two-sample t-tests were used to compare baseline characteristics and endpoints in the two study groups. Incident rates of diabetic nephropathy, retinopathy and neuropathy were calculated using the STATA (ir) procedure. Cox proportional hazards models were used to calculate HRs after dichotomizing the incidence of diabetic nephropathy, retinopathy, and neuropathy at each time point over the 10-year follow-up. We started with an unadjusted model. We then fitted a model with terms for five candidate covariates (sex, age, diabetes type, diabetes duration, and BMI) in addition to group assignment. Given the established sex disparities associated with incidence of diabetic nephropathy, we fitted a model with sex and group assignment only before fitting a fully adjusted model with the four remaining covariates. Data are expressed as mean (SD), mean (SEM) or n (%). All analyses were performed using STATA Special Edition V.15.0 for Windows (StataCorp, College Station, Texas, USA, 2017). In all tests, p<0.05 was considered statistically significant.