Drug titration patterns and HbA1c levels in type 2 diabetes

Objective: To evaluate oral antidiabetes drug (OAD) use, haemoglobin A(1c) (HbA(1c)) testing and glycaemic control in type 2 diabetes patients. Study design: Retrospective analysis based on claims data from the Integrated Healthcare Information Services (IHCIS) National Managed Care Benchmark Database. Methods: OAD use and HbA1c testing were analysed for patients with 2 claims indicating diagnosis of type 2 diabetes and >= 1 90-day OAD treatment period between 1 January, 2000 and 30 June, 2006. Likelihood of HbA(1c) testing was examined using multivariable logistic regression analyses, adjusting for OAD regimen and patients' sociodemographical characteristics. Results: Patients were classified based on initial OAD regimen: metformin (MET) (n = 22,203; 41.3%), sulphonylurea (SFU) (n = 18,439; 34.3%), thiazolidinedione (TZD) (n = 7663; 14.3%), SFU + MET (n = 5467; 10.2%) and TZD + MET (n = 2355; 4.2%). A total of 51.5% of patients had HbA1c testing during 90 days preceding OAD initiation through regimen completion. Approximately, 65% of MET and 58% of SFU patients had no titration of initial regimen. Patients demonstrating inadequate glucose control decreased from 68.5% at baseline to 46.9% within 90 days of regimen initiation. Multivariable logistic regression indicated several negative predictors of HbA1c testing, including SFU use, age 65+ years, moderate insurance copayment and preindex inpatient utilisation. Multivariable logistic regression of variables associated with reduced likelihood of up-titration included TZD, SFU + MET, or TZD + MET treatment, age 18-34 years, Medicare insurance and any preindex healthcare utilisation. Conclusions: Patients are not being transitioned to additional OADs in a stepwise fashion and/or are receiving inadequate titration on current OAD regimens. The low rate of HbA1c testing and rates of control are contributing factors.