This site is intended for health professionals only

Diabetes management and the glycaemic index

Lisa Vaughan
BAppSc GradDipNutrDietetics MSc
Diabetes Specialist Dietitian
Barts and the London NHS Trust
Hackney Diabetes Centre
Homerton University Hospital

Carbohydrates have traditionally been thought of in terms of simple and complex; however, the chemical nature of carbohydrates in foods does not completely describe their physiological effects.(1) The assumption that simple carbohydrates or sugars are quickly digested and absorbed, thus leading to poor glycaemic control, cannot be substantiated.(2,3) The glycaemic index (GI) classifies carbohydrates on the basis of their physiological effects, which is dependent on both the types of the constituent sugars and the physical form of the carbohydrate.(4)

The glycaemic index and carbohydrates
Dr David Jenkins and his co-workers developed the concept of the GI in 1981 as a method for classifying carbohydrate foods based on their glycaemic effect.(5) The research that has been carried out since then has challenged some of our widely held beliefs and found that the physiological responses to different carbo­hydrate foods are more complex than first thought.
The GI is a quantitative assessment, indexing the glycaemic response of a fixed amount of available carbohydrate in test foods compared with the standard food (either glucose or white bread). It shows, for example, that 50g of one carbohydrate food does not have the same effect on blood glucose as 50g of another ­carbohydrate food. The GI then ranks foods from 0 to 100 based on their glycaemic responses, providing an indication of the digestion rates of various carbohydrate foods.(4,6) Carbohydrate foods with a low GI are digested more slowly than those with a high GI and thus result in a reduction in both postprandial blood glucose levels and insulin response. While there is variability between how different individuals respond, the ranking always remains the same.
Carbohydrate foods have been shown to elicit a different glycaemic response based on a number of factors. Owing to this, the glycaemic response of carbohydrate foods is not able to be predicted by whether it is predominantly a sugar or a starch. Factors that influence the GI of a carbohydrate food include the amount of carbohydrate, type of sugar (glucose, fructose, sucrose, lactose), nature of the starch (amylase, amylopectin, resistant starch), cooking (degree of gelatinisation), food processing (particle size) and physical form of the food (such as the fibrous coat around beans), as well as other food components such as fat and protein, which can slow the digestion of the starch.(6-8)
It is important, however, that the concept of the GI should not be used in isolation, and foods should be neither included nor excluded on the basis of GI alone. The principles of a healthy, balanced diet should always guide food choices. Other considerations include local availability, cultural acceptability and the individual's preferences.(1)
The glycaemic load and carbohydrates
As the GI refers to a single food containing 50g of carbohydrate, it is important that we also take into account the fact that people may eat many carbohydrate-containing foods, in varying amounts, throughout the day. The glycaemic load (GL) represents both the quality of the carbohydrate food (the GI) as well as the quantity of that food (weight). Because of this, the GL is a better predictor of the impact of carbohydrate-containing food on postprandial blood glucose response and insulin secretion. Generally, foods with a high GI also have a high GL when consumed as a standard serving size. Some carbohydrate foods, however, can generate a high GL if they are dense in carbohydrates and/or eaten in large portions.(6,9)

The glycaemic index in diabetes
The major objectives in the dietary management of people with diabetes are to:(10)

  • Achieve a reduction in the risk of microvascular disease via normogylcaemia without undue risk of hypoglycaemia, especially in patients treated with sulphonylureas and/or insulin.
  • A reduction in the risk of macrovascular disease, including management of body weight, ­dyslipidaemia and hypertension.

Glycaemic effect
As low GI foods help to delay glucose absorption, they prevent large peaks in blood glucose, thus leading to a reduction in HbA(1c), as well as reducing peak insulin concentrations and overall insulin demand (see Figure 1).(4) Also, eaten at one meal, low GI foods will lower the GI of foods eaten at the subsequent meal. It is generally recommended that people aim for two low GI meals a day. When explaining the concept of the GI to patients, it is best to use the terms "slow" and "fast" carbohydrate foods: "low and slow" (see Table 1).



Management of hypoglycaemia
The treatment of hypoglycaemia is the exception to the low GI rule. For people with diabetes who are treated with either insulin or sulphonylureas, their blood glucose levels may sometimes fall below 4mmol/l. In this instance, it is important that the patient first takes 15g of a fast-acting, high GI carbohydrate, such as Lucozade, jelly beans or dextrose tablets (see Table 2). After 15 minutes, if the blood glucose levels have risen to above 4mmol/l, then the patient should either eat their usual meal or snack, if due within half an hour, or take some low GI carbohydrate, such as an apple, a yoghurt or a glass of milk (see Table 3). Step 1 should be repeated until the blood glucose level is greater than 4mmol/l.



The replacement of high GI foods for low GI choices at meals and for snacks can help to reduce or prevent the episodes of hypoglycaemia (including nocturnal hypoglycaemia) in people with diabetes.(4,10)
People with diabetes that take acarbose in conjunction with insulin or sulphonylureas must treat hypoglycaemia with glucose, not sucrose. This is because of its action as an alphaglucosidase inhibitor, preventing the hydrolysis of disaccharides. Acarbose used as sole therapy does not cause hypoglycaemia.

Weight management
Low GI diets also assist with weight management by prolonging the feeling of satiety (up to six hours, versus one hour for high GI foods), minimising postprandial insulin secretion (promoting fat versus carbohydrate oxidation), and maintaining insulin sensitivity. The postprandial effects of high GI, carbohydrate-dense foods may go some way towards explaining why low-fat diets have not always been effective when consumed in a free-living situation.(11,12)
Cardiovascular disease
The consumption of low GI/low GL diets is associated with a reduction in triglycerides and low-density lipoprotein (LDL) cholesterol, and a lower level of total to high-density lipoprotein (HDL) cholesterol. This positive effect on blood lipids suggests that these diets may be protective against developing heart disease as well as managing existing cardiovascular disease.(9)

Low GI foods are beneficial in the clinical management of diabetes by helping to achieve good glycaemic control as a result of minimising both hyperglycaemic and hypoglycaemic episodes, reducing the risk of cardiovascular disease and aiding weight loss. High and low GI foods more accurately reflect the physiological effect of carbohydrate foods than the traditional definition of "simple" and "complex". It is important, however, that both the type (GI) and quantity (weight) of the carbohydrate foods are taken into account in making recommendations, as well as continuing to ensure an overall nutritionally balanced diet based on the principles of the "Balance of Good Health".



  1. Food and Agriculture Organisation and World Health Organization. Carbohydrates in human nutrition - a summary of the joint FAO/WHO expert consultation. Geneva: WHO; 1997. Available from URL: http://www.sugar. ca/FAOPrt.htm
  2. Canadian Diabetes Association. Guidelines for the nutritional ­management of diabetes mellitus in the new millennium: a position statement by the Canadian Diabetes Association. Can J Diab Care 1999;23(3):56-69.
  3. Brand Miller JC. An introduction to the GI factor. In: Vaughan, L, editor. Current nutrition recommendations for adults with diabetes mellitus: a review. Unpublished. 1998. If you wish to access this paper please feel free to email the author.
  4. Augustin LS, Franceschi S, Jenkins DJA, Kendall CWC, La Vecchia C. Glycaemic index in chronic disease:a review. Eur J Clin Nutr 2002;56:1049:1-71.
  5. Ha TKK, Lean MEJ. Technical review: recommendations for the ­nutritional management of patients with diabetes mellitus. Eur J Clin Nutr 1998;52:467-81.
  6. Jenkins DJA, Kendall CWC, Augustin LSA, et al. Glycaemic index: overview of implications in health and disease. Am J Clin Nutr 2002;76 Suppl:266-73.
  7. Leeds A, Brand Miller J, Foster-Powell K, Colagiuri S. The glucose ­revolution. London: Hodder and Stoughton; 1996.
  8. American Diabetes Association. Evidence-based nutrition principles and recommendations for the ­treatment and prevention of diabetes and related complications [Position statement]. Diabetes Care 2002;25 Suppl:50-60.
  9. Bell S, Sears B. Low-glycaemic-load diets: impact on obesity and chronic diseases. Crit Rev Food Sci Nutr 2003;43(4):357.
  10. Willet W, Manson J, Liu S. Glycaemic index, glycaemic load, and risk of type 2 diabetes. Am J Clin Nutr 2002;76 Suppl:274-80.
  11. Brand-Miller J, Holt S, Pawlak D, McMillan J. Glycaemic index and obesity. Am J Clin Nutr 2002;76 Suppl: 281-5.
  12. Anderson H, Woodend D. Effect of glycaemic carbohydrates on short-term satiety and food intake. Nutr Rev 2003;61(5):S17.

Glycemic Index

BNF Information

Further reading
The New Glucose Revolution
Brand-Miller J, Leeds A. New York: Marlowe & Company; 2003.