Impact of oral antihyperglycemic therapy on all-cause mortality among patients with diabetes in the Veterans Health Administration.

Abstract

The objective of this analysis was to evaluate the impact of several classes of oral antihyperglycemic therapy relative to sulfonylurea monotherapy on all-cause mortality among a cohort of patients with diabetes from the Veterans Health Administration (VHA). A retrospective cohort study using data obtained from the VHA Diabetes Epidemiology Cohort was used. Users of oral antihyperglycemic therapy were classified into the following cohorts: sulfonylurea monotherapy, metformin monotherapy, metformin plus sulfonylurea, thiazolidinedione (TZD) use alone or in combination with other oral agents (TZD users), and no drug therapy. All-cause mortality was the outcome of interest. Multivariate mixed models incorporating a propensity score to account for imbalance among cohorts were used to estimate drug effects on mortality with associated 95% CIs. A total of 39,721 patients with diabetes were included in the study. Adjusted odds ratios and 95% CIs for all-cause mortality were 0.87 (0.68-1.10) for metformin monotherapy users, 0.92 (0.82-1.05) for metformin plus sulfonylurea users, and 1.04 (0.75-1.46) for TZD users, relative to sulfonylurea monotherapy users. We did not find any significant drug effect on all-cause mortality for any oral treatment cohorts relative to sulfonylurea oral monotherapy.