In this study we aimed to evaluate the effects of practice size (patient volume) and diabetes caseload in outpatient services on the quality of diabetes care in a teaching hospital. We analyzed the medical records of 2038 diabetic patients treated at a medical center in central Taiwan between January and June 2007. Outpatient practice size (including diabetic and non-diabetic patients) per clinic decreased the odds of glycated hemoglobin (A1C) testing (13% less for every 10 increase in outpatient encounters; p<0.001), and the percentage of A1C values<7% (8% less for every 10 increase in outpatient encounters; p=0.03) in diabetic patients treated by non-endocrinologists. However, a higher caseload of patients with diabetes was associated with an increased lipid profile measurement (19% more for every 5 increase in diabetic patients; p<0.001). In diabetic patients treated by endocrinologists, a higher patient volume was associated with increased odds of low-density-lipoprotein cholesterol (LDL-C) levels<100mg/dl, although there was no effect on the measurement and values of A1C. In conclusion, our study demonstrated some evidence of patient volume-outcome relationship in the management of diabetes in different specialties. This finding can have some implications to the health care system and the referral policy.