The promises of personalised nutrition: What is the science behind ZOE?

fruit and vegetables

Are personalised dieting regimes like ZOE any better than other forms of dieting? Photo credit: Dan Gold via Unsplash

Personalised nutrition programmes have recently emerged as a potentially more effective way for individuals to control their long-term health. One prominent player in this field is ZOE: a company that claims to go beyond conventional calorie counting. By integrating scientific data from various sources, ZOE aims to predict optimal personal nutrition, based on the claim that individuals have inherently different biological responses to different food groups.

ZOE’s approach relies on a multifaceted data collection process, involving a stool sample for microbiome bacteria analysis, a blood sample for blood fat analysis, and the use of a continuous glucose monitor (CGM). The CGM allows real-time tracking of postprandial (the state produced after eating a meal) blood glucose levels. A membership for ZOE costs £299.98 upfront, and an additional £24.99 each month.

ZOE claims to not be about dieting, but about adding foods into your diet.

ZOE uses a ‘ZOE score’ — a number from 1-100 that tells you how well your body responds to a food, instead of telling you how many calories or fats (etc.) are in the food. This is an alternative to ‘tell[ing] you what you can and can’t eat’. Furthermore, ZOE claims to not be about dieting, but about adding foods into your diet. The insights gained from the data collection process ‘allow individuals to make impactful changes to reduce dietary inflammation, improve gut health, weight, and overall health’.

The PREDICT studies, which comprise the data that ZOE collects, has found variability between individuals in postprandial responses, including changes in blood fats, glucose, and insulin levels. The ZOE programme argues this emphasises the need for personalised dietary guidance. Contrary to the notion that genetics plays a major role in shaping the microbiome, ZOE’s research indicates that even identical twins share only 34% of the same gut microbes. Therefore, ZOE uses the gut microbiome as a critical and modifiable target for personalised nutrition. Despite evidence that individual differences in responses to food intake do occur, the actual effect of this on overall health is debatable.

Personalised nutrition seems promising, but criticism from nutritional anthropologists highlights concerns about the lack of robust data that suggests programmes such as ZOE cause better weight loss than an otherwise healthy and varied diet. The PREDICT research concludes that ‘a diverse diet rich in minimally processed, high-fiber, plant-based foods supports the growth of healthy gut microbes’, and ‘a diet low in diversity and high in highly processed foods is associated with microbes linked to poor metabolic responses and long-term health.’ This insight, however, is not very personal, and is arguably common sense.

It is also unclear how insightful the use of a CGM to track blood glucose levels in individuals with normal levels is. CGMs are often used by diabetics. For example, the A1C test shows your average blood sugar level for the past two to three months. A normal A1C level is below 5.7 per cent. Eventually, ZOE may be able to provide insight into what healthy postprandial glucose levels actually are. However, it is important to remember that bodies without diabetes are designed to respond to blood sugar spikes and have the means to do so. When blood sugar levels rise, the pancreas produces the hormone insulin which allows body cells to absorb blood sugar for energy or storage. When blood sugar levels fall, the pancreas produces another hormone, glucagon, which signals to the liver to release stored sugar.

Therefore, non-diabetic users of CGMs are sometimes at risk of unnecessary health concerns when seeing a blood sugar spike on their monitor. A blood sugar spike, in itself, is not inherently unhealthy.

Companies like ZOE that use CGMs claim that glucose spikes are potentially dangerous, even in non-diabetics because of its association with inflammation. While inflammation can arise from metabolic stress, the evidence that glycaemic responses in people without diabetes causes inflammation is minimal.

Blood glucose levels are affected by a myriad of factors, not just food. For example, stress, sleep, the menstrual cycle, alcohol consumption, and body temperature can all effect glucose tolerance. Another significant factor which can affect the level your glucose ‘spikes’ after a meal is the ‘second meal effect’,  which is when the first meal eaten affects the subsequent meal’s glycaemic response. For example, breakfast carbohydrate tolerance is improved when foods with a low glycaemic index are eaten the previous evening.

Dr Nicola Guess argues that personalised nutrition programmes, like ZOE, ‘are taking what is scientifically plausible and prematurely presenting it as a panacea for weight loss or type 2 diabetes’. Experiences with Zoe vary, with some questioning its gamification aspect. For example, the use of ‘ZOE scores’ and the ‘very, very basic’ daily lessons provided by the app have been highlighted. One user commented that ‘it’s especially infuriating that it takes no account of the amount of activity you expend which feels extremely unscientific for a programme like this’.

Other users describe more positive outcomes, such as increased energy and improved well-being. Another user commented that they, ‘like[d] that no food is off limits and there is a focus on learning control and food combinations so you can still eat what you like or tweak recipes so that they score higher.’

However, when assessing the effectiveness of personalised nutrition programmes such as ZOE, it is important to consider the effect of dietary monitoring. The Hawthorne effect describes the phenomenon where observed people change their behaviour. Therefore, asking people to track what they eat can, of course, lead to changes in dietary intake. Statistics from Zoe’s studies, such as the fact that 70% of users experienced increased energy and 85% improved their gut health, are therefore not necessarily direct evidence of the effectiveness of personalised nutrition.

Personalised nutrition programmes are a long way off providing truly personal advice. 

Dietary advice should always be based on a person’s needs and preferences, but ZOE’s claims oversimplify what is an expansive, and growing, field of research. This is not to say there is not sound science behind the field of personalised nutrition, and that it may one day be a way for the average person to improve their health. However, the rush to commercialise personalised nutrition is premature, particularly when considering its high price-tag. Ultimately, personalised nutrition programmes are a long way off providing truly personal advice.