OBJECTIVES To describe the relationship between A1c levels and medication possession ratio (MPR) of three oral anti-hyperglycemic medications (sulfonylureas, biguanides and thiazolidinediones) for participants enrolled in a diabetes disease management program. METHODS This was a retrospective evaluation of participants enrolled in a diabetes disease management program. Continuously enrolled participants were included if there was a documented A1c value. The MPR was calculated from administrative pharmacy claims data. Participants were classified as low, moderate or high risk based on their A1c levels. ANOVA was used to analyze differences in mean MPR between the three groups. For participants on more than one drug class, a Pearson's correlation was used to analyze relationship of MPRs between different drug classes. RESULTS A total of 57 percent of participants on sulfonylurea therapy, 64 percent of participants on biguanides and 63 percent of those on thiazolidinediones reached an A1c goal of <7. Participants classified as high risk (A1c >9) had a lower MPR compared to participants in the low and moderate risk group (p<0.0001). For participants taking a sulfonylurea, the mean MPR for those who reached the predetermined A1c goal (<7) was 0.89 (±0.21) compared with 0.76 (±0.30) for those in the high risk group (p<0.001). MPR for thiazolidinediones was 15 percent lower in participants with A1c > 9 (p<0.001). Pearson correlation analysis showed a positive relationship between the MPRs in different drug classes. CONCLUSIONS Medication adherence as measured by the MPR was higher for participants who reached the target A1c goal of <7.