Findings presented today at the American Society of Nutrition and the American Diabetes Association conferences demonstrate how one-size-fits-all dietary guidelines are too simplistic, and that a personalised approach to nutrition is likely to provide better long-term health benefits.

The findings are a result of a study, involving more than 1,000 participants (mostly twins), which measured how blood levels of markers such as sugar, insulin and fat change in response to specific meals, along with data on activity, sleep, hunger and gut bacteria (microbiome) in thousands of participants in the US and UK.

The results reveal a wide variation in blood responses to the same meals, whether they contained carbohydrates or fat. For example, some participants had rapid and prolonged increases in blood sugar and insulin, which are linked to weight gain and diabetes. Others had fat levels that peaked and lingered in the bloodstream hours after a meal, raising the risk of developing heart disease.

These variations could only be partly explained by their genes - identical twins who share all their genes and experience the same environment often had different responses to identical foods.

The research was led by Professor Tim Spector and colleagues in the Faculty of Life Sciences & Medicine, as well as Massachusetts General Hospital and nutritional science company ZOE.

ZOE is using machine learning techniques to analyse this wealth of detailed nutritional data. They are developing a home-based test and an app that can measure and make predictions about an individual’s nutritional responses, helping people choose the best foods for them to manage their weight and overall health.

Professor Tim Spector, Professor of Genetic Epidemiology at King’s and lead researcher on the study said: “The sheer scale and detail of our scientific project is such that for the first time we can explore tremendously rich nutrition data at the level of an individual. Our results surprisingly show that we are all different in our response to such a basic input as food. It was a real shock to see that even identical twins have such different responses.”

The results suggest that personal differences in metabolism due to factors such as the gut microbiome, meal timing and exercise are just as important as the nutritional composition of foods, supporting the idea that simple nutritional labeling is insufficient for assessing food.

Dr Sarah Berry, Senior Lecturer in Nutritional Sciences at King’s and Scientific Advisor at ZOE explained: “For the first time, we're expanding large-scale nutritional research beyond blood sugar. These findings show that the responses to food of a number of key metabolic markers - including triglycerides, insulin and blood sugar - are highly individualised. No one has been able to combine data on this scale before.”