Differences between patients initiating insulin and exenatide in clinical practice in Greece (the CHOICE study)

OBJECTIVE Exenatide BID, in head-to-head Phase III clinical trials with insulin, provided similar glycemic control in patients whose diabetes mellitus (DM) was uncontrolled by oral antidiabetic medications (OADs). A variety of criteria appeared to influence the critical decision to move towards an injectable DM therapy. CHOICE is a multinational, prospective, non-interventional, observational study designed to assess the time to change to injectable glucoselowering therapy (exenatide or insulin) among adults with type 2 DM, and the factors associated with the choice. This paper describes baseline data from Greek patients in the study. METHOD In the course of routine clinical care, the demographic/clinical characteristics and healthcare resource use of patients initiating injectable antidiabetic therapy were analyzed, using univariate tests to quantify differences between cohorts on either exenatide or insulin. RESULTS Of 807 eligible patients (52.5% men, mean age 63 +/- 11 years), 318 (39.4%) initiated exenatide and 489 (60.6%) insulin, according to protocol-defined criteria. Patients initiating exenatide were younger than those initiating insulin (59 +/- 10 vs 65 +/- 11 years, p<0.0001), with a higher proportion of women than men (54.4% vs 42.9%, p<0.01), a higher mean body mass index (BMI; 34.4 +/- 7 vs 28.7 +/- 5 kg/m(2); p<0.0001) and waist circumference (112 +/- 15 vs 99 +/- 14 cm; p<0.0001). They also had lower mean levels of glycated hemoglobin (HbA1c) in the 3 months prior to initiation of injectable therapy (8.4 +/- 1.5% vs 9.3 +/- 1.9%, p<0.0001), and lower mean blood levels of low-density lipoprotein cholesterol, creatinine and both fasting and random glucose and less often had microalbuminuria. In addition, those initiating exenatide reported a shorter mean duration of DM (9 +/- 6 vs 12 +/- 8 years, p<0.001) with fewer macrovascular (21.4% vs 28.2%, p<0.05) or microvascular (10.4% vs 17.2%, p<0.01) complications. Of all the participants 45 (5.6%) reported >= 1 episode of hypoglycemia in the 3 months prior to initiation of injectable therapy, 15 (4.7%) initiating exenatide and 30 (6.1%) initiating insulin. More patients initiating exenatide had been given dietary and exercise advice (77.7% vs 68.9%, p<0.05). At the time of initiation of injectable therapy, 32% of patients initiated with insulin and 10% of those initiated with exenatide did not report taking any OAD therapy. CONCLUSIONS Patients initiating exenatide rather than insulin as injectable therapy for DM in Greece were on average younger and more obese, with lower HbA1c, a shorter duration of DM, and fewer macro-and microvascular comorbitities, while more had received diet and exercise instructions. The percentage of patients in the insulin group reporting no OAD use at the time of initiation of injectable therapy was 3 fold that in the exenatide group.