RT Journal Article
SR Electronic
T1 Estimating Effects Of Second Line Therapy For Type 2 Diabetes Mellitus: Retrospective Cohort Study
JF bioRxiv
FD Cold Spring Harbor Laboratory
SP 130724
DO 10.1101/130724
A1 Assaf Gottlieb
A1 Chen Yanover
A1 Amos Cahan
A1 Yaara Goldschmidt
YR 2017
UL http://biorxiv.org/content/early/2017/04/29/130724.abstract
AB Objective Metformin is the recommended initial drug treatment in type 2 Diabetes Mellitus, but there is no clearly preferred choice for an additional drug when indicated. We use electronic health records to infer the counterfactual drug effectiveness in reducing HbA1c levels and effect on body-mass index (BMI) of four second line diabetes drug classes.Study design and setting Retrospective analysis of the electronic health records of US-based patients in the Explorys database using causal inference methodology to adjust for censored patients and confounders.Participants and Exposures Our cohort consisted of roughly 25,000 patients with type 2 diabetes, prescribed metformin along with a drug out of four second line drug classes – sulfonylureas, thiazolidinediones, DPP-4 inhibitors and GLP-1 agonists, during the years 2000-2013.Main outcome measures Glycated hemoglobin (HbA1c) and BMI of these patients after six and twelve months of treatment.Results We show that all four drug classes reduce glycated hemoglobin levels, but the effect of sulfonylureas after 12 months of treatment is less pronounced compared to other classes. We also predict that thiazolidinediones increase body weight while DPP-4 inhibitors decrease it.Conclusion Our results are in line with current knowledge on second line drug effectiveness and effect on BMI. They demonstrate that causal inference from Electronic health records is an effective way for conducting multi-treatment causal inference studies.