OBJECTIVE: The goal of this study is to evaluate physicians' prescribing behavior based on the results of Hb1Ac tests for patients with type-2 diabetes. METHODS: The study design was retrospective covering the 12-month longitudinal period June, 2003 through May, 2004. We used a three-month look back period and a three-month look forward period for the validation of pre and post test (HbA1c) therapy. The study was based on the analysis of electronic CMS1500 claims for 33,764 unique patients (HIPAA compliant) and their results from the most recent HbA1c tests. Patients' medication was assessed through NCPDP electronic claims utilizing NDC codes and drug classes. Analysis included calculations of descriptive statistics and regression modeling. RESULTS: The high level of HbA1c was discovered during laboratory testing where a higher percentage of patients without medication were placed on medication. In particular, within a category of patients with HbA1c greater than or equal to 9mg/dL, this switch occurred 32.1% more frequently then within a category of patients with HbA1c less than 7mg/dL (p<0.01). For patients with HbA1c greater than or equal to 9mg/dL, the likelihood to change an existing therapy was 64.2% higher than for patients with HbA1c less than 7mg/dL (p<0.01). Also, for people with a high level of Hb1Ac, the likelihood to switch from monotherapy to polytherapy was 1.6 times higher than for people with a low level of testing. CONCLUSIONS: For patients with type-2 diabetes, a level of Hb1Ac is one of the important predictors of change in a physician's prescribing behavior. High levels of HbA1c most likely will cause a switch from just a dietary and exercise regimens to additional medication component.