The objective of the study was to assess the safety of changing ambulatory patients from animal insulin produced in the Czech Republic administered by classical insulin syringes to human insulins of the Danish firm Novo Nordisk, using a NovoPen 3 applicator. Furthermore antibody levels against hog, bovine and human insulin were assessed. Forty-seven patients with diabetes type I stabilized on an intensified insulin regime were after a four-day preparatory period divided at random into two groups. Patients in group A (n = 22) were after randomization changed to human insulin, patients in group B (n = 25) eight weeks later. From the onset of treatment with human insulins up to the end of the study the mean daily dose of insulin in both groups increased (in group A by 1.51 IU/day, in group 1.35 IU/day). This is not statistically or clinically significant. During the same period a statistically significant decline of the mean value of the daily 8-point glycaemic profile was recorded (in group A by 0.85 mmol/l, in group B by 0.51 mmol/l). Glycosylated haemoglobin declined also significantly in the course of the study (in group A by 1.64%, p = 0.00004, in group B by 1.02%, p = 0.0077). The greatest drop occurred during the preparatory period. Despite the increased daily insulin dose and improved compensation the number of hypoglycaemic events declined significantly in both groups (in group A by 0.78%, p = 0.0102, in group B by 0.74%, p = 0.0134). Hypoglycaemic coma was not recorded in either group. A significant drop of insulin antibodies was found in both group after the onset of treatment with human insulins. From the results of the study ensues that metabolically compensated type I diabetics with a mean daily insulin dose of 0.6 IU/kg body weight can be changed without any complications, in the ambulatory department, to human insulins with the same dosage. Concurrently a gradual decline of antibodies can be expected.