Group based diabetes self-management education compared to routine treatment for people with type 2 diabetes mellitus. A systematic review with meta-analysis

Table 5 Sensitivity analysis of the studies that had 12 months data on glycated haemoglobin (changes from baseline to 12 months) with comparison for intervention (Int) and control (contr) groups and the heterogeneity (measured by I ² ) of the analyses

Analysis number / Outcome	Effect measure	N studies	N participants (Int/contr)	Difference (95%CI)	P-value^A	Heterogeneity (I2)
5.1 Non-translated publications	Mean Diff	10	630/630	-0.52 (-0.75 to -0.28)	0.000	41.0
5.2 Number of participants more than median of all studies with 12 months HbA1c data(104 total)	Mean Diff	6	568/576	-0.67 (-0.86 to -0.48)	0.000	18.4
5.3 Studies with any part of results recalculated (HbA1c 12-14 months)	Mean Diff	6	450/438	-0.54 (-0.78 to -0.31)	0.000	0.0
5.4 Study quality (Low risk of bias):glycated haemoglobin 12-14 months	Mean Diff	2	192/187	-0.42 (-1.00 to 0.17)	0.163	77.4
5.5 Study quality (Moderate risk of bias)	Mean Diff	7	428/436	-0.55 (-0.79 to -0.31)	0.000	29.7
5.6 Study quality (High risk of bias)	Mean Diff	2	130/130	-0.95 (-1.55 to -0.35)	0.002	0.0
5.7 Study quality (Low & Moderate risk of bias)	Mean Diff	9	620/623	-0.51 (-0.72 to -0.29)	0.000	41.3
5.8 Drop-out rate less than 10 %	Mean Diff	5	444/454	-0.79 (-0.97 to -0.61)	0.000	0.0

A. The P- value is calculated for the difference between the intervention and control group.