Seven RCTs (n=1,687) were included in the review.
Twice-daily premixed insulin analogue regimens versus once-daily basal insulin analogue regimens (3 RCTs).
With regard to glycaemic control, the change in HbA1c levels ranged from -1.00 (standard deviation, SD, 0.85) to -2.79 (SD=0.11) percentage points for premixed insulin and from -0.42 (SD=0.92) to -2.36 (SD=0.11) percentage points for basal insulin; the results in each study were statistically significant (p<0.01).
Hypoglycaemia rates, adjusted per 1 year, ranged from 3.4 (SD=6.6) to 8.2 (SD=16.8) for premixed insulin and from 0.7 (SD=2.0) to 5.4 (SD=13.0) for basal insulin; the results in each study were statistically significant (p<0.05).
The total daily insulin dose ranged from 0.42 (SD=0.2) to 0.82 (SD=0.4) U/kg for premixed insulin and from 0.36 (SD=0.18) to 0.57 (SD=0.37) U/kg for basal insulin; the results in each study were statistically significant (p<0.05).
Three-times-daily premixed insulin analogue regimens versus once-daily basal insulin analogue regimens (3 RCTs).
With regard to glycaemic control, the change in HbA1c levels ranged from -0.72 (SD=0.89) to -1.2 (SD=1.1) percentage points for premixed insulin and from -0.3 (SD=1.1) to -0.75 (SD=0.1) percentage points for basal insulin; the results in each study were statistically significant (p<0.01).
Hypoglycaemia rates (2 RCTs) ranged from 0.7 (SD=1.7) to 4.7 (SD=6.35) for premixed insulin and from 0.3 (SD=0.8) to 2.31 (SD=3.24) for basal insulin; the results were statistically significant for both studies (p<0.05).
The total daily insulin dose ranged from 0.35 (SD=0.26) to 0.7 (SD=0.3) U/kg for premixed insulin and from 0.28 (SD=0.21) to 0.6 (SD=0.3) U/kg for basal insulin; the results were statistically significant (p<=0.01) (2 RCTs).
Twice-daily premixed insulin analogue regimens versus intensive basal bolus insulin therapy (1 RCT).
With regard to glycaemic control, the change in HbA1c levels was -1.56 (SD not reported) percentage points for premixed insulin and -1.23 percentage points for intensive basal bolus insulin; the results were statistically significant (p<0.01).
Minor hypoglycaemia was noted in about 30% of the patients in both treatment arms, but severe cases were noted in only 5% of the patients receiving intensive basal bolus insulin.
The total daily insulin dose was 0.86 U/kg for premixed insulin and 0.63 U/kg for intensive basal bolus insulin.
Other results (percentage of patients achieving HbA1c target, fasting blood glucose and change in body weight) were reported. Confidence intervals of the results were not reported.