The broad objective of this work is to examine the relationship between diabetes medications and bladder cancer. The prevalence of diabetes is increasing as our population ages and becomes more obese. The population health impact of this trend is magnified by the potential for diabetes to increase the risk of cancer. The association between diabetes and cancer is well established but poorly understood. Both basic science and epidemiologic evidence suggests anti-diabetic drugs may contribute to increased cancer risk among diabetics, but key questions remain. Bladder cancer is the 6th most common cancer in the United States. There is a strong body of evidence linking diabetes to bladder cancer and our own work has linked oral diabetes treatments in general to bladder cancer. The development and diffusion of diabetes management models and treatment guidelines, largely promoting intense use of pharmaceuticals, has resulted in high rates of use of single and, more commonly, combined diabetes treatment regimens creating substantial diabetes medication exposure in a growing population of diabetic patients. To our knowledge, no studies have specifically examined the cancer promoting effects of combined diabetes therapy in general and in relation to bladder cancer specifically. We propose to address this knowledge gap by studying a cohort of more than 800,000 Medicare beneficiaries with type 2 diabetes. In this cohort, we anticipate approximately 3,200 incident bladder cancer cases over a 5-year period. This study will examine the association between diabetes treatments (individual and combined drugs) and bladder cancer. The objective of Aim 1 is to determine drug exposure patterns among Medicare beneficiaries with type 2 diabetes.
In Aim 2, we compare the incidence of bladder cancer among users of distinct anti-diabetes therapies. This will include direct comparisons of oral agents and comparison of insulin therapies (human and analog).
Our third Aim i s to conduct an exploratory analysis of the association of common anti-diabetic drug combinations with bladder cancer, using appropriate type 1 error spending methods. Time-to-event analysis including Cox's model, time-dependent covariates, inverse propensity weighting and classification and regression trees will be employed. This work will inform clinical practice and shared decision making. It will permit development of models, datasets, and expertise applicable to future studies of the effect of diverse pharmaceuticals on a broader range of cancer types.

Public Health Relevance

Research suggests diabetes treatment drugs may increase the risk of cancer for people using them, but the evidence is not clear and most research is limited to assessment of just one drug at a time. Many patients with diabetes take multiple medications concurrently to control their disease. To improve understanding of the possible relationship between diabetes treatment and cancer risk we will study rates of bladder cancer associated with diabetes medication use. Our study will examine bladder cancer associated with use of individual medications as well as bladder cancer associated with concurrent use of multiple medications.

National Institute of Health (NIH)
National Cancer Institute (NCI)
Small Research Grants (R03)
Project #
Application #
Study Section
Special Emphasis Panel (ZCA1)
Program Officer
Filipski, Kelly
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
Dartmouth College
Public Health & Prev Medicine
Schools of Medicine
United States
Zip Code