GI diets don’t work – gut bacteria and dark chocolate are a better bet for losing weight
The mainstays of most of the diet regimens of the last 30 years have been the GI (glycaemic index) rating score as well as its cousin the glycaemic load. Famous best-selling diet books such as the G-Plan Diet, the South Beach diet all used the index in some way and changed the way we thought about carbohydrates. Now a detailed new study published in Cell pays this score – and how we use it – some closer scrutiny.
The GI theory goes that there are many different types of carbs and they can be graded into how rapidly the body converts them into glucose. The faster the burn rate, the higher the index and the more rapid the rise in blood sugar. This surge in blood sugar also triggers a rise in insulin and the combination of these events if sustained over time is believed to lead to unhealthy metabolic changes leading eventually to obesity and diabetes.
This nutritional dogma has been the backbone of the advice to avoid eating high GI foods such as pasta, rice and potatoes and replace them with low GI alternatives such as beans or lentils. In many cases people give up carbs completely or avoid refined carbs and instead go for approaches like the Atkins Diet. This cutting down on “bad” carbs also sounds like common sense and most doctors, health professionals and members of the public, including myself when I started writing my book The Diet Myth, assume that proper scientific and clinical trials have been carried out that back up these ideas. But what if none of it were true and we had been misled?
When I started researching four years ago I could find plenty of laboratory animal studies showing the benefits of a low GI diet. But while studies of diabetic subjects generally confirmed some advantage of a low GI diet, in non-diseased people it was a different story. There was a distinct lack of any decent human trial that proved convincingly that a GI-based diet was any better than equivalent diets based on cutting calories.
While studies did show that low GI foods and diets could alter blood sugar profiles no study has shown that when the amount of carbs and calories are kept constant the lower GI score for the food didn’t influence body weight. So humans seem to be responding differently to laboratory mice, and although many people do lose weight in the short-term on these diets, it may be because they are paying more attention to what they eat rather than any validity of the food scoring process.
It’s those microbes again
What the new study – from an Israeli team led by Eran Segal – found was that by closely studying 800 volunteers who kept careful food diaries and had wearable devices that continually monitored their blood glucose levels, an amazing pattern of diversity emerged.Pacific Northwest National Laboratory, CC BY-NC-SA
People eating the same foods and had the same GI scores had very different glucose responses. Some of this was due to existing weight and age differences in the study’s subjects – but differences remained even in similar people that they couldn’t explain. That was until they looked inside their intestines at the thousands of species of microbes we have inside us. We all have around 100 trillion bacteria mainly living in our colons (outnumbering our cells by ten to one) and unlike our genetic code, which is 99.9% similar across individuals, we probably only share less than 20% of our microbes on average.
Our gut microbes live off the fibre in foods that make it to the lower part of the intestines. The researchers found that the highly variable microbial profile in the guts of the volunteers determined how quickly the food was broken down and the rate at which glucose appeared in the blood. This was a much stronger effect than the type of carb being eaten. Some people could eat potatoes without any surge in blood sugar and others with the wrong set of microbes just had to look at one and their blood levels peaked.
Unlike our genes, our microbes are to some extent changeable. The team went on to successfully modify the diets of the subjects with the worst looking profiles.
These results could be a game-changer for nutrition. We can now move from our outdated obsession with inaccurate calorie counts and GI scores of food items to working out how our foods interact with our microbes.
We should focus on the fibre type and content of each item and assess its ability to act as a fertiliser for microbes via chemicals called polyphenols. For example, purple coloured potatoes have three times more polyphenols than those with normal skins and appear healthier in humans.Potatoes by Shutterstock
White rice lacks polyphenols so is unhealthy while dark chocolate has plenty and doesn’t cause a glucose surge. This could also be a game changer for the millions suffering from or (if overweight) at high risk of diabetes.
Rather than avoiding all foods which might be harmful and reducing your fibre and diversity, profiling your gut microbes from a piece of toilet paper could tell you whether you should be eating potatoes or not. By using the breakthrough techniques of gene sequencing we can now accurately and cheaply identify the different patterns of microbes. While the Israeli team are working on a commercial testing kit – others can now join crowdfunded research projects in the US and the UK that want people to be able to test themselves and properly map and explore the amazing diversity of our gut microbes. This is the exciting new era of personalised nutrition and for many people a jacket potato could be back on the menu.
Tim Spector is affiliated with the crowd funded American Gut Project and is director of the related British Gut microbiome project. He is author of a book on gut microbes: 'The Diet Myth': the secret behind what we eat, by W&N 2015
Tim Spector is a Professor of Genetic Epidemiology at Kings College, London & Director of the Department of Twin Research and Genetic Epidemiology at St Thomas’ Hospital, London. Professor Spector graduated from St Bartholomew’s Hospital Medical School, London. After working in General Medicine, he completed a MSc in Epidemiology, and his MD thesis at the University of London.
He founded the UK Twins Registry of 11,000 twins in 1993, which is one of the largest collections of genotype and phenotype information on twins worldwide. Its breadth of research has expanded to cover a wide range of common complex traits many of which were previously thought to be mainly due to ageing and environment. He has published over 700 research articles on common diseases and is ranked in the top 1% of world scientists.
He has written several original articles on the heritability of a wide range of diseases and traits including back pain, acne, inflammation, obesity, memory, musical ability and sexuality. He has published widely on obesity, food and nutrition. He also is interested in new areas of biology such as epigenetics and recently our gut microbiome and is director of the British Gut project
He has written several books, He is also author of - The Diet Myth: The real science behind what we eat by W&N 2015 and Identically different: Why you can change your genes, by W&N in 2012 and Your Genes Unzipped in 2003.