Characteristics of Patients with Type 2 Diabetes Mellitus Receiving Treatment with Dapagliflozin Versus Sitagliptin: An Analysis of Commercial Claims and Linked Laboratory Data

Rationale: In order to compare the real-world use of commonly prescribed second-line oral diabetes therapies, real-world data comparing patients receiving treatment with dapagliflozin vs sitagliptin are needed, and limited resource use and cost data exist for patients initiating dapagliflozin in the real-world setting. Objectives: Primary: The primary objective of this study is to compare real-world health care resource utilization and costs following treatment with dapagliflozin versus sitagliptin. Secondary: The secondary objectives of this study are to assess clinical outcomes and treatment patterns among patients receiving treatment with dapagliflozin versus sitagliptin. Study Design: Retrospective cohort study. Target Subject Population: Patients receiving treatment with either dapagliflozin or sitagliptin will be evaluated. Study Variable(s): Primary Variables: Health care resource utilization and costs Secondary Variables: Demographics measured at the index date, Quan-Charlson Comorbidity score, measured during the baseline period, Agency for Healthcare Research and Quality (AHRQ)-based comorbidity measures, receipt of other antidiabetic medication classes during the baseline and follow-up periods, number and percentage of patients with a diagnosis of obesity during the baseline or follow-up periods, treatment patterns during the follow-up period (i.e., index dose, duration of treatment, discontinuation, adherence, receipt of additional antidiabetic medication classes), HbA1c outcomes (among subgroup of patients with linked laboratory data, sample size permitting), renal impairment, hypoglycemia as defined by a claims-based algorithm. Statistical Methods: Initial analyses will be descriptive in nature and entail the tabular display of mean values, medians, ranges, and standard deviations of continuous variables of interest (e.g., patient age) and frequency distributions for categorical variables (e.g., sex, geographic location). Outcomes will be compared between patients receiving dapagliflozin versus sitagliptin using univariate tests. Propensity score matching will be undertaken to reduce bias in the comparison of patients receiving treatment with dapagliflozin versus sitagliptin. Following matching, demographics and baseline characteristics will be assessed using standardized differences to determine balance in the post-matched sample. Outcomes will be assessed using tests for paired data (paired t-tests, signed rank tests, McNemar’s tests).

Rationale: In order to compare the real-world use of commonly prescribed second-line oral diabetes therapies, real-world data comparing patients receiving treatment with dapagliflozin versus sitagliptin are needed, and limited resource use and cost data exist for patients initiating dapagliflozin in the real-world setting. Objectives: Primary: The primary objective is to compare real-world health care resource utilization and costs following treatment with dapagliflozin versus sitagliptin. Secondary: The secondary objectives of this study are to assess the following real-world data among patients receiving treatment with dapagliflozin versus sitagliptin: 1. Patient demographic and baseline clinical and treatment characteristics 2.Treatment patterns, specifically dose at index date, duration of treatment, adherence, discontinuation, and receipt of additional antidiabetic agents 3. Clinical outcomes after treatment augmentation, including hypoglycemic events, renal impairment, and HbA1c Study Design: Retrospective cohort study of commercial medical and pharmacy claims along with enrollment information and linked laboratory results between July 1, 2013 and April 30, 2016. Target Subject Population: Patients receiving treatment with either dapagliflozin or sitagliptin will be evaluated in this analysis. The date of the first observed claim for either dapagliflozin or sitagliptin will define the index date. This study will use an intent- to-treat population, and patients will remain in their defined study cohort for the duration of the follow-up period, regardless of treatment changes. The baseline period will be defined as the 6 months’ pre-index date, and the follow-up period will be defined as the 12 months’ post-index date (inclusive of the index date). Patients will be required to have continuous health plan enrollment during the baseline and follow-up periods. Patients having at least one claim with a diagnosis of type 1 diabetes mellitus in the baseline period or gestational diabetes or evidence of pregnancy in the baseline or follow-up periods will be excluded from the study population. Furthermore, patients will be excluded from the analysis if they received a selective sodium glucose co-transporter-2 (SGLT-2) or dipeptidyl peptidase 4 (DPP-4) inhibitor during the baseline period. Patients receiving treatment with dapagliflozin will be matched on demographic, clinical, and economic characteristics to patients receiving treatment with sitagliptin using the 1:1 propensity score matching technique. Study Variable(s): Primary Variables: Health care resource utilization and costs Secondary Variables: 1. Demographics measured at the index date 2. Quan-Charlson Comorbidity score, measured during the baseline period 3. Agency for Healthcare Research and Quality (AHRQ)-based comorbidity measures 4. Receipt of other antidiabetic medication classes during the baseline and follow-up periods. 5. Number and percentage of patients with a diagnosis of obesity during the baseline or follow-up periods 6. Treatment patterns during the follow-up period 7. HbA1c outcomes 8. Renal impairment 9. Hypoglycemia as defined by a claims-based algorithm Statistical Methods: All analyses including the assessment of patient demographics, patient characteristics, and health care utilization and costs will be descriptive in nature and entail the tabular display of mean values, medians, ranges, and standard deviations of continuous variables of interest and frequency distributions for categorical variables. Prior to matching, outcomes will be compared between patients receiving dapagliflozin versus sitagliptin using univariate tests. In Phase 2, propensity score matching will be undertaken to reduce bias in the comparison of patients receiving treatment with dapagliflozin versus sitagliptin. Following matching, demographics and baseline characteristics will be assessed using standardized differences to determine balance in the post-matched sample. Outcomes will be assessed using tests for paired data.