During the early stages of the disease, most diabetes patients receive metformin to control high glucose levels associated with presence of diabetes. However, as the condition progresses, additional medications (i.e., on top of metformin) may be used. The type of additional medication used depends on a number of factors including (but not limited to) their effects on patients' kidney function - a key complication associated with diabetes progression. One such class of medications called dipeptidylpeptidase-4 (DPP-4) inhibitors have been thought to improve kidney function among people with diabetes. By comparison another commonly used class of medications called the sulfonylureas (SU) do not appear to have such benefits. However, we are aware of only one clinical trial and one real world study that compared the effects of these two classes of medications on outcomes related to kidney functioning directly. These studies had very few patients using a particular type of medication from each class and the outcomes were compared over short periods of time. Our study expands on this prior research by comparing the kidney-related outcomes of all diabetes patients treated with DPP-4 inhibitors to those treated with SU (both in combination with metformin) in the United Kingdom (UK) after 2007.
The proposed study will assess the comparative effectiveness of using DPP-4 inhibitors vs. SU in addition to metformin in 2007-2014 in terms of incidence of renal outcomes (albuminuria, doubling of serum creatinine, changes in estimated glomerular filtration rate) among T2DM patients with more than 3 months of metformin alone in the UK. In addition, the study aims to describe the characteristics of patients receiving DPP-4 inhibitors vs. SU after more than 3 months of metformin therapy, with the earliest indications of simultaneous use of background metformin and either DPP-4 inhibitor or SU considered as the index date. Baseline covariates will be descriptively compared between the two cohorts using chi-squared tests for proportions and Wilcoxon rank-sum tests for continuous measures. Outcomes will be assessed during at least 12 months after the index date. Time to first occurrence of the outcomes of interest will be described using Kaplan-Meier survival analyses and log-rank p-values. Multivariate Cox proportional hazard models will be used to assess the differences in incidence rates adjusting for baseline differences in demographics (e.g., age, gender), clinical attributes (e.g., renal function, glycemic burden), comorbidities, and other medication use. Sensitivity analyses will estimate K-M curves and Cox models for propensity-score matched cohorts with similar baseline characteristics.
Primary outcome: Incident microalbuminuria Other outcomes: Change in baseline category of albuminuria Change in serum creatinine level from baseline Stages of kidney disease as assessed using eGFR.
Noam Kirson - Chief Investigator - Analysis Group, Inc.
Urvi Desai - Corresponding Applicant - Analysis Group, Inc.
Christopher Edmonds - Collaborator - Astra Zeneca Inc - USA
Hung Heong Teh - Collaborator - Astra Zeneca Inc - USA
Jayanti Mukherjee - Collaborator - Bristol-Myers Squibb - USA ( BMS )
Kamlesh Khunti - Collaborator - University of Leicester
Katherine Tsai - Collaborator - MedImmune
Mark Meiselbach - Collaborator - Analysis Group, Inc.
Michael Hellstern - Collaborator - Analysis Group, Inc.
Phillip Hunt - Collaborator - Astra Zeneca Inc - USA
Zitong (Bruce) Jia - Collaborator - Analysis Group, Inc.