Type 2 diabetes (T2DM) is a chronic condition characterized by elevated blood sugar levels (hyperglycaemia) that can result in an increased risk of a variety of conditions including heart disease, strokes, kidney failure, blindness and amputation. Whilst initially patients may control their blood sugar by lifestyle modification (diet and exercise), ultimately most will require therapeutic intervention with regimens that increase in complexity as T2DM progresses. Exenatide is a relatively recent anti-diabetic drug which is known to lead to weight loss as well as improved blood glucose control. It has also been associated with reduced heart attacks and strokes. In this study we wish to use the CPRD database to compare outcomes for patients prescribed exenatide QW (Bydureon) compared with those prescribed basal insulin; a more established treatment for T2DM. In particular we wish to look at blood sugar control and weight change as the primary outcomes but also consider other outcomes: death, heart attacks and strokes. As the choice to treat with exenatide QW or insulin will be related to patient characteristics which may in themselves be associated with the outcomes of the study we aim to match study patients on some of these key variables and adjust for others in our analysis.
We wish to compare outcomes for T2DM patients treated with therapeutic regimens involving exenatide QW with those involving basal insulin-based therapies. T2DM patients will be selected from the CPRD GOLD, HES and ONS datasets. Those identified as being initiated on the target therapies between 2009 and 2014 will be extracted and their characteristics described. We will then match exenatide QW patients (cases) to basal insulin patients (controls) on key characteristics. For the primary objectives we wish to compare change in HbA1c and change in weight independently and as combined endpoints based on patients achieving a fall in HbA1c below 7.0% accompanied with a) any weight loss and b) weight loss of >/= 5% of baseline. Secondary objectives are to compare cost of primary and secondary care resource use, time to all-cause mortality and time to cardiovascular events. Times to endpoints will be compared using Cox proportional hazards models adjusting for demographic and clinical covariates. Change in weight and HbA1c will be compared using the Wilcoxon signed rank test. Health care costs will be estimated from standard tariffs and compared using the Mann-Whitney U-test. All analyses will be replicated based on patients prescribed exenatide (Byetta).
Craig Currie - Chief Investigator - Cardiff University
Bethan Jones - Corresponding Applicant - Pharmatelligence Limited t/a Human Data Sciences
Kristina Johnsson - Collaborator - Astra Zeneca Inc - USA
Qing Qiao - Collaborator - AstraZeneca Ltd - UK Headquarters
Sara Jenkins-Jones - Collaborator - Pharmatelligence Limited t/a Human Data Sciences
Susan S Grandy - Collaborator - Astra Zeneca Inc - USA
HES Admitted Patient Care;ONS Death Registration Data