Type 1 Diabetes: Efficacy of Carbohydrate Counting

MedicalResearch.com Interview with:
Kirstie Bell
Diabetes Dietitian, CDE & PhD Candidate
Human Nutrition Unit
The University of Sydney

MedicalResearch.com: What are the main findings of the study?

Answer: Overall, the evidence to support carbohydrate counting is limited, with current data showing a non-significant improvement in HbA1c.  Pooled results from 7 quality randomised control trials studies showed carbohydrate counting had no significant effect on glycemic control (-0.35%, p = 0.096).  There was a significant improvement in HbA1c of 0.64% points in studies in adults that were conducted in a parallel design.

This is the first meta-analysis of carbohydrate counting in type 1 diabetes. Up until now, it has not been known what improvement in glycemic control can be expected. Current international guidelines for diabetes management have been based merely on gradings of the available evidence. However, assessing the overall effectiveness of carbohydrate counting is critical in clinical practice to guide medical and dietary management decisions.
MedicalResearch.com: Were any of the findings unexpected?

Answer: It was surprising how few quality, randomised controlled trials were published.  Carbohydrate counting is considered the gold-standard method of adjusting mealtime insulin doses yet this meta-analysis shows the scarcity of high-level evidence to support this recommendation.

MedicalResearch.com: What should clinicians and patients take away from your report?

Answer: Overall, there is limited evidence to recommend carbohydrate counting over other dietary interventions for improving glycemic control in type 1 diabetes.

This does not mean we should abandon carbohydrate counting in practice.  This is a well-established intervention and is still the best method known for matching insulin to meals.

The non-significant result does however highlight that there are some limitations to the efficacy of carbohydrate counting in practice and that clinicians and patients may not be able to expect ideal glycemic control even with perfect carbohydrate counting.

MedicalResearch.com: What recommendations do you have for future research as a result of this study?

Answer: This study emphasises the need additional evidence to support carbohydrate counting in practice.  This would provide the opportunity to further improve clinical practice by investigating better ways of doing things.  For example, there are different methods of counting carbohydrates including counting in grams, exchanges and portions.  Additional studies using these different methods, will allow analysis of whether any method is superior to the others and thus further improve clinical practice.

There is also a lack of evidence for carbohydrate counting in children and adolescents.  We could only identify 1 randomised, controlled study in this area yet carbohydrate counting is widely used in this population.

This meta-analysis also opens the doors to looking at alternative methods for determining mealtime insulin doses.  Recent studies from around the world have been examining the effect fat and protein has on insulin requirements and whether mealtime insulin dosing needs to be expanded to encompass more than just carbohydrate counting. This meta-analysis provides a benchmark for comparing carbohydrate counting with these new dietary interventions.


Efficacy of carbohydrate counting in type 1 diabetes: a systematic review and meta-analysis
Kirstine J Bell BNutrDiet,Alan W Barclay PhD,Peter Petocz PhD,Stephen Colagiuri FRACP,Prof Jennie C Brand-Miller PhD
The Lancet Diabetes & Endocrinology – 1 February 2014 ( Vol. 2, Issue 2, Pages 133-140 )
DOI: 10.1016/S2213-8587(13)70144-X

Last Updated on February 12, 2014 by Marie Benz MD FAAD