Update On Carbohydrate Recommendations for Type 2 Diabetes

Some people are calling for low-carbohydrate diets to be recommended for people with diabetes but what does the evidence show? Dr Alan Barclay APD, GI Foundation Chief Scientific Officer, looks at the issue.

Type 2 diabetes is increasing rapidly around the globe, due to a large number of factors including our ageing population, decreased physical activity and changes to the foods we eat and drink, amongst many others. While all factors are undoubtedly important, food and nutrition seem to gain a disproportionate amount of attention and generate the most debate. Hardly a week goes by without a new “diet” that self-proclaims it will help people prevent or manage their diabetes better than all preceding diets. Food faddism not only creates confusion and angst for people with diabetes, it also inadvertently affects their families, friends and carers who they share their food with. Low carbohydrate diets are becoming the latest fad, having last been popular in the 1970’s.

To help put healthy eating into perspective, the American Diabetes Association(1) has crafted a number of food and nutrition management goals that summarise all of the main points that need to be considered about healthy eating for diabetes:

1) Achieve and maintain

* Blood glucose levels in the recommended range or as close to recommended as is safely possible

* Blood pressure levels in the recommended range or as close to recommended as is safely possible

* A blood cholesterol and triglyceride profile that reduces the risk of vascular (e.g., heart attack, stroke, retinopathy, kidney and peripheral vascular) diseases

2) Prevent, or at least slow, the rate of development of the chronic complications of diabetes by modifying food and nutrient intakes and lifestyle (e.g., physical activity, drinking, sleep, smoking habits, etc…)

3) Consider individual nutrition needs, taking into account personal and cultural preferences and willingness to make changes

4) Maintain the pleasure of eating by only limiting food choices when indicated by scientific evidence

While points 1 and 2 are self-evident and the focus of most health professionals that work with people with diabetes, points 3 and 4 are arguably just as important, as ultimately they determine how well an individual will be adhere to a dietary regimen in the long-term.

What scientific evidence is there for recommending low carbohydrate diets over other dietary patterns? To answer this, we first need to define what a low carbohydrate diet is. The following definitions have recently been recommended(2):

  • Very low-carbohydrate diets provide 20-50 g/carbohydrate/day or less than 10% of a 8,400 kJ diet.
  • Low-carbohydrate diets provide less than 130 g/carbohydrate/day or less than 26% of energy from a 8,400 kJ diet.
  • Moderate-carbohydrate diets provide 130-230 g/carbohydrate/day, or 26 – 45% of energy from a 8,400 kJ diet.
  • High-carbohydrate diets provide greater than 230 g/carbohydrate/day or 45% of energy from a 8,400 kJ diet.

It is also important to consider how much carbohydrate the average Australian, and person with diabetes, is eating at present. The most recent national nutrition survey(3) estimates that the average adult consumed 222 g/carbohydrate/day in 2011/12, and the most recent study of people with diabetes(4) estimates that they consumed 214 g/carbohydrate/day in the 1990’s. In other words, the average Australian consumes a moderate amount of carbohydrate, and it’s likely that people with diabetes consume less.

The American Diabetes Association(1), Canadian Diabetes Association(5) and Diabetes UK(6) have all recently conducted independent systematic reviews of the scientific evidence for the management of diabetes and are all in agreement that there is no evidence for strict macronutrient recommendations for people with diabetes. Instead, there are a variety of healthy eating patterns that may be suitable, based on the individual’s personal and cultural preferences and willingness to change. These may include, but are not limited to Mediterranean style, vegetarian/vegan, low-fat and low-carbohydrate patterns of eating(1).

Systematic reviews and meta-analyses(7) have shown that over the medium-term, all of these different patterns of eating can lead to decreases in HbA1c between 0.12-0.47% points, with low-carbohydrate diets being the least effective (0.12% reduction) and Mediterranean style (0.47% reduction) and vegetarian (0.39% reduction)(8) being the most effective. Unfortunately no long-term (> 2 years) studies have been published to-date.

Since these systematic reviews were conducted, Australia’s CSIRO investigated the effect of a very low-carbohydrate (50 g / day), high fibre (24.7 g / day) and low saturated fat diet versus a moderate carbohydrate (205 g / day), low GI diet in people with diabetes over 12 months(9). Both patterns achieved similar decreases in body weight (~10 kg), BMI (~3.3 kg/m2), waist circumference (~9.5 cm), fasting glucose (-0.7 (low carb) vs -1.5, P=0.1), HbA1c (-1% in both), BP, total and LDL cholesterol. Medication score, glycemic variability, HDL (0.1 vs 0.06 mmol/L) and TG (-0.4 vs -0.01 mmol/L) improved modestly on the very low CHO diet compared to the moderate carbohydrate diet. These results would therefore not materially affect the conclusions of the earlier systematic reviews.

Dietary patterns with widely differing carbohydrate composition may produce similar outcomes due to differences in glycemic index (GI), glycemic load (GL) and insulin responses. It’s important to note that while the amount and type of carbohydrate are the most powerful predictors of the effect of foods on blood glucose(10) and insulin levels(11), protein and fat also have an independent effect(12;13).

Glycemic index values are frequently found on the labels of carbohydrate-containing foods in Australia (e.g. foods with the GI Symbol (see below) and online and total carbohydrate is a mandatory component of the Nutrition Information panel, so people with diabetes can easily calculate the glycemic load (GL = GI ÍCarbohydrate per serve ÷ 100) of commonly consumed foods and beverages.

The best way to use the GI is to swap from regular high GI to low GI alternatives within a food group, and this approach also typically identifies the lower GL choices. The insulin response of a food is highly correlated with its glycemic load(14) and for common foods and meals may be used as a surrogate at present. Table 1 lists commonly consumed low GI foods in Australia.

Table 1: Low GI foods

Dense wholegrain bread
Authentic sourdough bread
Pasta and noodles
Pearl cous cous
Doongara rice
Traditional oats
Natural muesli
Most fruits and vegies (except melons and most potatoes)

Reducing the amount of carbohydrate is therefore simply one method of reducing blood glucose and insulin levels, but it is not the only way, nor necessarily the best way – that depends on the individual. Every person with diabetes should see an Accredited Practicing Dietitian for personalised advice.


  1. Evert AB, Boucher JL, Cypress M, Dunbar SA, Franz MJ, Mayer-Davis EJ, et al. Nutrition therapy recommendations for the management of adults with diabetes. Diabetes Care 2014 Jan;37 Suppl 1:S120-43. doi: 10.2337/dc14-S120.:S120-S143.
  2. Feinman RD, Pogozelski WK, Astrup A, Bernstein RK, Fine EJ, Westman EC, et al. Dietary carbohydrate restriction as the first approach in diabetes management: critical review and evidence base. Nutrition 2015 Jan;31(1):1-13.
  3. Australian Bureau of Statistics. Australian Health Survey: Nutrition First Results – Foods and Nutrients, 2011–12 — Australia.  9-5-2014. Canberra, Australia, Australian Bureau of Statistics.
  4. Barclay AW, Brand-Miller JC, Mitchell P. Macronutrient intake, glycaemic index and glycaemic load of older Australian subjects with and without diabetes: baseline data from the Blue Mountains Eye study. Br J Nutr 2006 Jul;96(1):117-23.
  5. Canadian Diabetes Association Clinical Guidelines Expert Committee. Canadian Diabetes Association 2003 clinical practice guidelines for the prevention and management of diabetes in Canada. Canadian Journal of Diabetes 2003;27(Suppl 2):S1-S152.
  6. Diabetes UK. Evidence-Based Nutrition Guidelines for the prevention and management of diabetes. 2011.
  7. Ajala O, English P, Pinkney J. Systematic review and meta-analysis of different dietary approaches to the management of type 2 diabetes. Am J Clin Nutr 2013 Mar;97(3):505-16.
  8. Yokoyama Y, Barnard ND, Levin SM, Watanabe M. Vegetarian diets and glycemic control in diabetes: a systematic review and meta-analysis. Cardiovasc Diagn Ther 2014 Oct;4(5):373-82.
  9. Tay J, Luscombe-Marsh ND, Thompson CH, Noakes M, Buckley JD, Wittert GA, et al. Comparison of low- and high-carbohydrate diets for type 2 diabetes management: a randomized trial. Am J Clin Nutr 2015 Oct;102(4):780-90.
  10. Wolever TM, Bolognesi C. Prediction of glucose and insulin responses of normal subjects after consuming mixed meals varying in energy, protein, fat, carbohydrate and glycemic index. J Nutr 1996 Nov;126(11):2807-12.
  11. Bao J, Atkinson F, Petocz P, Willett WC, Brand-Miller JC. Prediction of postprandial glycemia and insulinemia in lean, young, healthy adults: glycemic load compared with carbohydrate content alone. Am J Clin Nutr 2011 May;93(5):984-96.
  12. Pal S, Ellis V. The acute effects of four protein meals on insulin, glucose, appetite and energy intake in lean men. Br J Nutr 2010 Oct;104(8):1241-8.
  13. Wolpert HA, Atakov-Castillo A, Smith SA, Steil GM. Dietary fat acutely increases glucose concentrations and insulin requirements in patients with type 1 diabetes: implications for carbohydrate-based bolus dose calculation and intensive diabetes management. Diabetes Care 2013 Apr;36(4):810-6.
  14. Bao J, de J, V, Atkinson F, Petocz P, Brand-Miller JC. Food insulin index: physiologic basis for predicting insulin demand evoked by composite meals. Am J Clin Nutr 2009 Oct;90(4):986-92.

Pin It on Pinterest

Share This