Fifteen RCTs (n=3,270 participants, range 23 to 689) were included.
Self-monitoring was associated with significantly lower HbA1c than no monitoring (WMD -0.31%, 95% CI -0.44 to -0.17, I2=33%; 12 RCTs). There was little evidence of publication bias. Sensitivity analyses yielded no statistically significant differences between parameters. Self-monitoring significantly increased the chance of detecting hypoglycaemia (RR 2.10, 95% CI 1.37 to 3.22, I2=59%; six RCTs).
There was no significant difference in HbA1c between the groups for frequent versus less frequent self monitoring (four RCTs, I2=79%).