Like many of you, I was shocked to wake up recently to headlines describing the results of a Monash University study that purported to show that a low carb diet led to an increase in the diagnosis of type 2 diabetes.

As this was contrary to virtually every other published study, not to mention the hundreds of thousands of people who have improved their blood glucose management and even put their diabetes into remission with a low-carb diet, I decided to do a deep dive into the study.

About the study

The study, titled “Association of low carbohydrate diet score with the risk of type 2 diabetes in an Australian population: A longitudinal study,” was authored by a group of researchers from Monash University led by Professor Barbora de Courten and published in the journal Diabetes & Metabolic Syndrome: Clinical Research & Reviews.

The researchers described their method: “Between 1990 and 1994, the Melbourne Collaborative Cohort Study (MCCS) recruited 41,513 people aged 40–69 years. The first and second follow-ups were conducted in 1995–1998 and 2003–2007, respectively. They analysed data from 39,185 participants. The Low Carbohydrate Score (LCS) was calculated at baseline as the percentage of energy from carbohydrates, fat, and protein. The higher the score, the less percentage of carbohydrates contributed to energy intake. They then divided the subjects into five groups (quintiles) based on their LCS from lowest to highest. They then compared the numbers who developed type 2 diabetes at the two time points between the different quintiles and concluded that “LCD score was associated with increased risk of T2D.”

This study has so many problems that it is hard to know where to start, but let me take you through some of the major issues.

Defining a low carb diet

First and foremost, they did not study a low carb diet at all. The lowest quintile had 37.5% of total energy input from carbs—no one would consider this low carb, which is defined by most as less than 26%, and many argue even lower. So, they are comparing high carb with very high carb.

The “low carb“ quintile 5, with a carb intake of 37.5% compared to 55.4% in quintile 1, had increased fat content from 28.7% to 41.5% – an increase of 12.8%. However, only 2.8% of that was saturated fat. The remainder was other types of fat likely to be largely omega-6 polyunsaturated and trans fats, given the study timeline and the massive increase in processed and ultra-processed food intake during that time. Omega-6 and trans fats have been shown to increase the incidence of obesity and type 2 diabetes.

Inconsistencies in data

There are numerous methodological flaws in the paper.

First, the paper claims that the subjects were categorised into five EQUAL groups based on their low carbohydrate scores. Yet, there are enormous discrepancies in the number of participants in each group (Quintile 1 = 9134, Quintile 5 =7637). In addition, the “low carb” quintile was more socio-economically disadvantaged, had more former or current smokers, more former or current drinkers, was less physically active, and had higher measures of obesity (BMI) and central obesity (waist-to-hip ratio) than participants in the highest carb quintile. All these factors are risk factors for the development of type 2 diabetes.

In addition, many of the numbers and percentages quoted do not add up, and the authors have confused kilojoules of energy intake with kilocalories, which is a pretty basic error.

The link between LCD scores and type 2 diabetes

But, the biggest deception in the whole paper revolves around the number who developed diabetes in the second wave, which was approximately 13 years after baseline. The paper clearly states that there is NO significant association between the incidence of diabetes and the LCD score quintiles in Wave 2, despite all the additional risk factors in that group.

So how can they then claim that “LCD score was associated with increased risk of T2D”? I am not sure.

The issue of obesity

The final point is that any differences they observed disappeared completely once they took into account obesity (BMI – 76% of the difference) and central obesity (waist-to-hip ratio).

This is a poor study, and to be honest, I’m not sure how it got published. Any reviewer should have been able to see the numerous flaws.

I think we can safely ignore this study and be guided by the numerous studies which show the exact opposite – that a low carb diet is very beneficial for those with type 2 diabetes.