Clinical Outcomes of Exenatide Versus Basal Insulin

Objectives of the study To compare changes in HbA1c and weight from baseline in patients initiated exenatide based-therapy regimen with those initiated insulin-based therapy regimen among type 2 diabetes patients who were injectable naïve in primary care in UK. Health care utilization will be calculated for each of the treatment cohorts, and compared.

Data source Patients will be selected from the Clinical Practice Research Datalink (CPRD), a longitudinal, anonymized research database derived from nearly 700 primary-care practices in the UK. These practices are considered representative of the UK as a whole. The primary-care dataset (CPRD GOLD) comprises data on demographics, diagnoses, hospital referrals, prescriptions emanating in primary care, and other aspects of patient care. Approximately 60% of practices participate in a linkage scheme, by which their patient records are linked to other data sources, including the Hospital Episode Statistics (HES) dataset which provides data on all inpatient and outpatient contacts occurring within National Health Service hospitals in England, and the Office for National Statistics (ONS) mortality dataset. Diagnostic information in the CPRD primary-care dataset is recorded using the Read code classification a UK primary-care practice standard. HES inpatient data are recorded using the International Classification of Disease codes, 10th revision (ICD-10); ONS mortality data are recorded using the ICD-10 and ICD-9 classifications.

Patient identification and Method

Patients defined by CPRD as being of acceptable research quality will be classified as having type 2 diabetes if they had a Read code indicative of diabetes and at least one of the following selection criteria applies:

1. more than one diagnostic record exclusively for type 2 diabetes,
2. prescriptions for two or more different classes of non-insulin glucose-lowering therapy,
3. a diagnostic code indicative of type 2 diabetes (regardless of conflicting diagnoses of type 1 or nonspecific diabetes) plus a prescription for a non-insulin glucose lowering therapy.

Two groups of cases will be defined as patients receiving exenatide as monotherapy or in combination with one or more other oral glucose-lowering therapies based on therapy codes recorded within CPRD. The groups will be:

1. exenatide once weekly formulation (Bydureon),
2. exenatide twice daily formulation (Byetta). Patients will be excluded if they had received

1. prior injectable diabetes therapy 2. less than 90 days continuous exposure to the exenatide therapy Control patients will be selected from the pool of type 2 diabetes patients who had a prescription for basal insulin as monotherapy or in combination with one or more other oral glucose-lowering therapies. The same exclusion criteria as that applied to cases was applied, that is

1. prior injectable diabetes therapy
2. less than 90 days continuous exposure to basal insulin

Matching will be performed by direct matching and by propensity score matching at a ratio of 1:1 based on the following criteria:

1. age (±5 years)
2. gender,
3. year of index exposure (±1 year),
4. diabetes duration (±2 years),
5. BMI (±3 kg/),
6. HbA1c [±1% (±11 mmol/mol)
7. Concurrent glucose-lowering medication.

Index date is the initiation of the study drugs between January 2009 and December 2014.

Primary care and secondary contacts will be ascertained from the CPRD GOLD and HES databases respectively. Admissions will be described by number, length of stay and cost. Healthcare resource groups (HRGs) will be assigned to each patient spell and processed using HRG 4 grouper software (National Casemix Office, Winchester, UK). The allocated HRG will then linked to the 2014 National Tariff adjusted for nature of the admission (elective admissions versus emergency) and excess length of stay. Primary care costs were derived from the Units costs of Health and Social Care 2014 Baseline characteristics of patients initiated on either Bydureon/Byetta and basal insulin will be presented for all patients and for those included in the direct matched and propensity score matched analyses. Differences between characteristics will be tested using the t-test for continuous variables and Pearson chi-square test for categorical variables.

Outcomes All patients received standard care, no interventions were given by the study investigator.

The primary outcome will be HbA1c and weight change. Baseline measures will be defined as any measurement between -180 days and baseline. Change will be measured from baseline to 6 months and 12 - 24 months (± 90 days) for those patients remaining on their index regimen.

HbA1c change from baseline at 6 months and 12-24 months will be calculated and differences between treatment groups compared. The proportion of patients for whom HbA1c falls below 7.0% will also be compared.

Weight change from baseline at 6 months and 12-24 months will be calculated and differences between treatment groups compared. Differences will considered as both an absolute and relative change from baseline.

Two composite end points will be considered based on the proportion of patients:

1. reaching a target of HbA1c<=7.0%% with weight reduction
2. reaching a target of HbA1c<=7.0%% with weight reduction ≥5% The proportion of patients reaching these endpoints were compared by chi-square test. Analyses will be performed for all patients and those matched directly and matched by propensity score.

Secondary outcomes - Health service utilisation Rates and costs of health service contacts will be calculated and compared between treatment groups using the Mann-Whitney U-test. Follow up will be defined as time from index date to last prescription date + 90 days.