We analyzed data from 42 institutes using CoDiC^®, an electronic diabetes data collection and management system, to determine insulin injection regimen effectiveness and rapid-acting insulin analog (RA) efficacy on glycemic control in those with type 2 diabetes mellitus (T2DM). Out 22,307 subjects with T2DM registered in 2006, we analyzed data from 1,094 treated with insulin for 6 months or more. During this period, HbA₁c decreased in those undergoing two injections/day of a 30% RA (30RA) or 30% regular insulin (30R) preparation-a decrease greater in those treated with 30RA. HbA₁c markedly decreased in those undergoing prandial boluses of RA or R insulin. Basal-bolus insulin with RA or R insulin plus NPH insulin or long-acting insulin analog (LA) dramatically decreased HbA₁c, particularly in those given RA. This shows that prandial and basal-bolus insulin maintain glycemic control more effective than two injections/day of mixed insulin preparations. Better glycemic control was attained with RA insulin than with R insulin.