[Conversion of fast insulin intensive therapy to lispro insulin in type I diabetes. Pharmacoeconomic analysis of cost-effectiveness].

In order to analyze the initial cost-effectiveness of transfer to two treatments with insulin lispro in type 1 diabetes, a pharmaco-economic study was conducted for nine months. After an educational reinforcement, a group of 30 C-peptide-negative patients (31.8 +/- 11.5 years [mean +/- SD], time since diagnosis of diabetes of 9.2 +/- 7.1 years, and on intensive therapy for 5.3 +/- 3.1 years) initiated a 3-month basal period with their usual therapy (preprandial rapid insulin and nocturnal NPH). Patients were then randomly assigned to one of the two groups, changing rapid insulin to either lispro (L1) or lispro combined with 15% to 20% NPH insulin (L2). Cross-over was made 3 months after the first treatment. Efficacy and safety were established by the assessment of HbA1c, self-monitoring blood glucose and hypoglycemia rates. Therapy cost was measured by systematic examination of the injection devices and wastage of insulin. The mean prescribed and actually consumed doses for R, L1, L2 groups were 52.9, 57.1, 55.2 U and 60.3, 64.1, 63 U per day, respectively (p < 0.001). The average of postprandial peak glucose (9.7, 8.4, 8.3 mM; p < 0.001) and HbA1c (7.6%, 7.2%, 7.1%; p < 0.01) were significantly lower after L1 or L2 lispro therapy. Although no statistical differences in overall hypoglycemia rates were observed, fewer nocturnal episodes were detected (0.72, 0.37, 0.41 events/month). The mean daily cost for regular insulin treatment was lower (186.8, 241.8; 215.7 pts and 53.7 pts per day. Efficacy and safety for two MIT regimens containing lispro were similar in the short run. Nevertheless, the preprandial use of the fast-acting insulin analog lispro in combination with a 15%-20% intermediate-acting NPH seemed to be more cost-effective than the premeal lispro therapy alone.