Design: Systematic review and meta-analysis of randomised trials
Population: 11 trials, 1,096 adults with type 2 diabetes
Primary outcome: Fear of hypoglycaemia score
Effect: SMD −1.89 (95% CI −2.54 to −1.25); I² 95%; prediction interval −4.34 to 0.56

A meta-analysis pooled 11 randomised trials of structured education or behavioural programmes in 1,096 adults with type 2 diabetes, looking at fear of hypoglycaemia.

The pooled standardised mean difference was −1.89 (95% CI −2.54 to −1.25), which would be a very large effect. But heterogeneity was extreme (I² 95%), the prediction interval ran from −4.34 to 0.56, and there were signs that small positive studies were over-represented. GRADE certainty was very low.

Fear of hypoglycaemia drives under-dosing of insulin and sulfonylureas, missed meals and deliberate high glucose, so it is worth asking about. This review supports addressing it but cannot say which programme works, or how well.

Ask patients on insulin or sulfonylureas directly whether fear of lows changes how they dose or eat
Teach recognition and treatment of hypoglycaemia as part of routine education
Review regimens that cause frequent lows before blaming the patient's anxiety
Consider CGM alarms where affordable for patients whose fear is limiting treatment

Why it matters

Hidden fear of lows is a common, fixable reason for poor control that rarely gets asked about.

The statistics, in plain English

An I² of 95% means the trials disagree far more than chance would explain. The prediction interval crossing zero means a new programme in a new setting might show no benefit at all.