Efficacy of carbohydrate counting in type 1 diabetes: a systematic review and meta-analysis

Background Although carbohydrate counting is the recommended dietary strategy for achieving glycaemic control in people with type 1 diabetes, the advice is based on narrative review and grading of the available evidence. We aimed to assess by systematic review and meta-analysis the efficacy of carbohydrate counting on glycaemic control in adults and children with type 1 diabetes. Methods We screened and assessed randomised controlled trials of interventions longer than 3 months that compared carbohydrate counting with general or alternate dietary advice in adults and children with type 1 diabetes. Change in glycated haemoglobin (HbA(1c)) concentration was the primary outcome. The results of clinically and statistically homogenous studies were pooled and meta-analysed using the random-effects model to provide estimates of the efficacy of carbohydrate counting. Findings We identified seven eligible trials, of 311 potentially relevant studies, comprising 599 adults and 104 children with type 1 diabetes. Study quality score averaged 7.6 out of 13. Overall there was no significant improvement in HbA(1c) concentration with carbohydrate counting versus the control or usual care (-0.35% [-3.9 mmol/mol], 95% CI -0.75 to 0.06; p=0.096). We identified significant heterogeneity between studies, which was potentially related to differences in study design. In the five studies in adults with a parallel design, there was a 0 64% point (7.0 mmol/mol) reduction in HbA(1c) with carbohydrate counting versus control (95% CI -0.91 to -0.37; p<0.0001). Interpretation There is some evidence to support the recommendation of carbohydrate counting over alternate advice or usual care in adults with type 1 diabetes. Additional studies are needed to support promotion of carbohydrate counting over other methods of matching insulin dose to food intake.