Main Edition
US
FR
Professor Tim Spector was one of the “winners” of the Covid era: his ZOE symptom tracker app accrued millions of users during the pandemic.
Now he has pivoted back to his true passion, gut health, and taken many of his followers with him. Endorsed by celebrities such as Davina McCall and Carrie Johnson, the new version of the ZOE app promises a personalised nutrition plan and comes with a glucose blood monitor usually used by diabetics. It is proving hugely popular, with over 100,000 subscribers paying up to £600 in their first year — and a further 300,000 on the waiting list.
It boasts all the hallmarks of a scientific endeavour, with endorsements by world-leading experts and numerous studies. But how convincing are its claims?
Deborah Cohen, a medically qualified TV, print and radio reporter, and Margaret McCartney, a GP, undertook a forensic investigation for UnHerd and found that ZOE’s scientific foundations aren’t as strong as they would have you think…
Read on to learn about:
1. ZOE and the influencers
2. Why Tim Spector created ZOE
3. Does ZOE think you are healthy?
4. Why ZOE wants your data
5. What does your gut tell ZOE?
6. How scientific is ZOE?
***
View this post on Instagram
“I gave up after a month.” Gail, 49, is talking about her experience of using the ZOE app — the heavily promoted personalised nutrition plan that, we’re told, will help us understand our bodies and make us more healthy. “It certainly didn’t change my food compulsion — if anything, it made me feel worse.”
Take a look at ZOE’s lavish series of adverts, or the rapturous response to the app on social media, and you would be forgiven for thinking that Gail is an anomaly. TikTok, Facebook and Instagram are filled with celebrities, ZOE “members” and influencers sharing their positive experiences and inviting you to join the club.
One is Carrie Johnson, who recently shared her welcome email on Instagram, another is Davina McCall, one of the faces of ZOE’s marketing campaign. She claims she is now “living her best life”, telling prospective consumers (and her 1.7 million Instagram followers) that the knowledge provided by the technology is the “greatest gift ever”. A vocal menopause-awareness campaigner, McCall claims the gut microbiome is “completely different” in pre- and post-menopausal women.
But women only make up part of the market. Steven Bartlett, of Diary of a CEO podcast, is also an ambassador: “There’s an incredible correlation between what I’ve eaten, my blood glucose levels, and how I’m feeling,” he claims.
These three are among the 117,000 people who ZOE claims are now “members”. Another 300,000 are on the waiting list. This, despite the fact that ZOE isn’t cheap: it costs £24.99 per month (roughly £300 per year) and £299.99 for the intro kit which includes the instantly recognisable bright yellow blood glucose monitor that users display proudly.
These monitors are all part of a “personalised nutrition” plan that also measures your gut “microbiome”: the collection of “good” and potentially “bad” microorganisms in your intestines. This is assessed via stool samples which you collect in a “hammock” over the loo bowl and then post to the “lab” for testing.
After a fortnight of continuous blood sugar monitoring, combined with your gut bug measurement, you are provided with a nutritional programme — basically foods you should eat and those you should avoid. ZOE suggests that this sort of personalised approach can help with energy levels, reduce hunger, help you reach a healthy weight, feel less bloated, sleep better, and avoid chronic health issues.
But is ZOE the solution? Certainly, its stratospheric rise is a marketing triumph. But beneath the sheen of its glossy adverts and its wealth of medical testimonies, ZOE’s scientific foundations seem far from healthy…
2. Why Tim Spector created ZOE
From its inception in 2018, ZOE has relied heavily on the profile and academic credentials of co-founder Tim Spector, professor of genetic epidemiology at King’s College London. Already a familiar name in nutrition circles, he rose to prominence during the pandemic with his ZOE Covid symptom tracker app, described by The Financial Times as “arguably the world’s largest ever citizen science experiment”. With millions logging on to share their test results and symptoms, Spector’s app, boosted by £5 million of government funding, was credited with first highlighting that loss of taste and smell accompany Covid infection.
A self-described “measured risk-taker”, Spector told The FT that he moves from subject to subject “if I’m not very interested in the next step or the next stage any more”. He started out as a rheumatologist, treating joint and muscle disorders, before turning to epigenetics, the study of how behaviours and environment can affect the way your genes work. Then, he set up the UK Twins Registry, one of the richest collections of data about identical and non-identical twins in the world.
It was this work, in particular, that led to his current interest: the interplay between diet and gut health. When he was studying the genetics of obesity in twins, he noticed that their weight sometimes greatly varied — which he attributed to the fact that their gut microbes were different.
After this Eureka moment, he decided to focus his efforts on nutrition, writing several books, the most recent being Food for Life: The New Science of Eating Well (2022). And he takes his own advice seriously — eating plenty of “fermented food” such as kombucha, kefir and kimchi to help his microbiome. He also microdoses on a diabetes drug which, he claims, could have anti-ageing benefits.
According to Spector’s website, it was during one of his presentations on his microbiome research that he met his ZOE co-founders, the now-CEO Jonathan Wolf and president George Hadjigeorgiou, both of whom have backgrounds in machine learning and business. They “shared his vision of helping individuals understand their unique biology to improve their own wellbeing in an accessible way”.
After opening offices in Boston and London, they rapidly attracted $27 million in initial funding to commercialise Spector’s work. (He remains a part-time ZOE employee and company shareholder.) And in 2019, they paid for a study that seemed to confirm the key concept for ZOE’s entry point into a crowded consumer market of products that offer advice on the healthiest foods to eat.
The PREDICT1 study suggested individuals’ metabolic responses to the same foods vary remarkably, with genes only playing a small part. (More on what you can infer from it later.) At a conference in the US, Spector revealed that they had found “really surprising” variation between certain biomarker responses — including some of those used by ZOE — to standardised meals.
It was what other investors needed to hear. The combination of a personalised nutrition app coupled with machine-learning and big data proved to be alluring buzzwords for the wellness-obsessed digital world. Steven Bartlett invested more than £2 million earlier this year. ZOE, he says, is a company that represents “the future”. It is “health science, driven by big data” that doesn’t just extend “our lifespan, but more importantly our health span”. Venture capitalists and those involved in a crowdfunding exercise thought similarly. Last year, one of the CEOs valued the company at over £209 million.
Yet some are sceptical. They suggest this is just the latest iteration of Big Diet — or Big Nutrition. “This is a classic playbook. X doesn’t work; the experts are wrong; read my book and find out the truth. Then buy my product which FINALLY fixes everything,” says one critic, who did not want to be named. Interestingly, Spector himself used to be of these sceptical voices, writing cautiously in The British Medical Journal in 2018 that observational studies — such as those since published by ZOE — are limited by the inability to measure causal relations between microbes and health traits. They can only show associations.
“The strongest level of evidence is obtained from interventional clinical studies — in particular, randomised controlled trials,” he wrote. (ZOE is yet to publish controlled trial data.) In 2019, he again suggested evidence was limited. In the same year he promoted ZOE study findings at that US nutrition conference, Spector wrote in a medical journal, that “there is a lack of long term human studies, or indeed follow-ups of short-term dietary interventions” to understand whether “diet-induced modulation of the gut microbiota endures”. Could this barrier be overcome?
3. Does Zoe think you are healthy?
On social media, it doesn’t take long to find people posting about the thrill of receiving their ZOE test results. As one user, Joel, 28, said to us: “It gives you another thing to share on Facebook or Twitter.” But how much can we actually infer about what’s going on inside our bodies?
There has long been a temptation in medical circles to assume that the more biological parameters we measure, the better. The truth, however, is that not all test results are useful. Just because something is measurable in your body, doesn’t necessarily mean that the result will yield useful information.
Dr Guess claims the app might be doing little more than recommending more fruit and veg to everyone. “They have no published data showing all this info results in personalised dietary advice,” she says. “They need to show that the dietary recommendations made by their algorithm can be replicated by other scientists. In other words, does the test give the same dietary recommendations on Wednesday as it did on Monday? This is one of the hallmarks of good science. Even more important, they need to show what they’re doing actually improves a user’s short and long-term health.”
This isn’t to say that continuous glucose monitoring isn’t useful. It is offered in the NHS, but for specific reasons — for people with diabetes. Readings can help to adjust insulin doses and improve control of potentially harmful spikes and troughs. Many people say that it has revolutionised their care. In people without diabetes, however, blood glucose is under continuous physiological control — with the pancreas regulating the release and storage of sugar in the body to keep it within normal parameters. This is the market ZOE is aiming for.
They’re not the only company making continuous glucose monitors aimed at people without diabetes. Even those in the clinical medicine arena are pivoting into this space. Indeed, The Global Wellness Institute estimates the global market segment for “healthy eating, nutrition and weight loss” is worth $946 billion a year.
The only problem is we don’t really have any evidence that these monitors do keep non-diabetic people healthy.
Dr Shivani Misra, a consultant in metabolic medicine at Imperial College London who researches continuous glucose monitoring, cautions that we don’t totally understand how to interpret measurements in healthy people. “We have no robust evidence base to suggest that those rises and falls are signalling anything untoward in the present or that they have negative consequences in the future,” she says. In fact, different devices might have different readings in response to eating the same meal.
Dr Kevin Hall investigates the regulation of body weight and metabolism for the publicly funded US National Institutes of Health, which has a programme in precision nutrition. His team has studied the use of continuous glucose monitoring in highly controlled environments — when people actually live at the lab for a month and undergo very specific diets and exercise regimens. They measured the glucose levels of one person wearing two devices. They found that sometimes one of the devices would give a very high glucose reading after a meal, and the other would give a low reading. They also analysed readings on the same device. They changed the regimens week to week — and concluded that the response to the same meal might vary as much as it did to eating a different meal in different weeks.
“Even in these very highly regimented kinds of controlled feeding studies, they don’t give rise to reproducible measures,” Dr Hall says. “It’s very likely that if that person ate a banana the next week, even if they had exactly the same previous food and did the same exercise that day, they might get a very different response.”
Unsurprisingly, the results are causing confusion. For instance, Gail says her glucose levels went up when she was exercising and her friend, who has diabetes and is used to monitoring their glucose, looked at her graph and thought it looked high. “I felt I didn’t get enough explanation about what this means or what might happen to my glucose,” Gail says. On TikTok, users also warn against the unwanted “spike”, saying they need to get a flatter line and describing their worries about their glucose dipping at night.
But this is not necessarily something to worry about: glucose going up and down is a normal bodily response. “Any time they see a slight rise or spike, people panic and it’s creating unnecessary anxiety,” Dr Misra says. “Then they think it’s wrong to eat carbohydrates — which release glucose into the blood — and there is no evidence that this is necessary or beneficial. Some tell me it’s taken over their life.”
But what about those who do feel their energy levels drop and notice a lethargic “crash”? Joel, for example, frequently wakes up at 3am to find his glucose dipping. He was even told he was waking up because of this dip. But he’s not entirely sure what to do about it.
This isn’t entirely surprising. While symptoms and how we feel are real, it can be hard to attribute cause and effect. In her clinic, Dr Misra sees people with and without diabetes who tell her they’re getting symptoms of low blood sugar, but when they’ve actually measured it, it’s within normal range. “There’s a risk everything becomes glucose-related,” Dr Misra says. “It’s human nature when you’re monitoring something you become hypervigilant about that.” This is something Dr Hall and his team also found in their studies. “I think the fascinating thing is that when people are provided with more and more information, they tend to make these correlations,” he says.
The risk is that we are creating problems for people rather than solving them, and turning healthy people into patients.
Ironically, perhaps mitigating this is the relationship between social class and health: in other words, the people most likely to be able to afford ZOE may be among the least likely to benefit from it. Because of this, Amitava Banerjee, professor of clinical data science at University College London and consultant cardiologist, says that ZOE plays into what is called the “digital divide”.
“Even when digital interventions are useful, the people who need them least are usually accessing them most, especially given the cost,” he says. “Those who are at the highest risk of health problems are often in more deprived and ethnically diverse areas, with neither the money nor the resources to use ZOE.” And even though it’s been largely tested on young healthy people of European ancestry — as the ZOE team has written in medical journals — the marketing goes beyond this cohort of people. “The results they get from studies might not be applicable to people outside that group,” Professor Bannerjee cautions.
Elsewhere, some doctors have turned to social media to complain about the increased workload the app is creating for them among a cohort known as the “worried well”. GPs have described to us how they’ve been contacted by patients concerned they have diabetes, as they think that is what their continuous glucose monitors are showing. While ZOE is just one of dozens of private companies and influencers who generate blood results capable of causing alarm and sending patients to NHS GPs, they all add pressure on appointments — especially if people didn’t get good enough information about what the test results actually mean.
5. What does your gut tell ZOE?
Blood sugar is only a part of what ZOE tests. Other indicators as to the function of our metabolism and how we respond to food is determined by our gut microbiome. Zoe wants to harness the “potential power” of this to affect your long-term health and certain symptoms, and so takes a reading on how good your “gut bugs” are with the help of your stool sample. A diet plan to improve that is then constructed.
But not everyone is convinced there is sufficient scientific evidence to justify this approach. Dr Hall says research suggests you can change the microbiome in pretty consistent ways across individuals by changing their diet. “The complexity of the microbiome kind of belies a single readout of good versus bad. How you even come up with a single good, bad kind of criteria is mystifying to me,” he says.
Professor Tariq Iqbal, a gastroenterology consultant who runs trials into faecal microbiota transplantation in people who have bowel disorders, says that we’re only just beginning to understand to what degree levels of microbes naturally fluctuate in the gut in a healthy person over time. “At the moment, there is no reason to get your microbiome generally tested,” he says. “We don’t have good evidence about what a “normal” microbiome looks like. It’s more complicated than simply saying there are ‘bad’ and ‘good’ bacteria in the gut.”
Current research suggests that there is no “standard” microbiome composition that can be used as a baseline for health. There may also be problems with reading too much from a single sample. According to Guts UK, a digestive disease charity that works closely with the Gut Microbiome for Health Group of the British Society of Gastroenterology, the bacteria found in your stool does not necessarily tell you what microbes are in your gut. Meanwhile, the British Dietetic Association warns: “Commercial gut microbiota testing and lifestyle change programmes based on the test results cannot be advised currently as the science is not at a stage that can support testing and treatments programmes.”
Nor, Dr Guess adds, is there any robust evidence to support the claim that personalising a diet based on the gut microbiome is more effective than generic healthy eating advice for any of the claims being made. In fact, Professor Iqbal says that patients might be unnecessarily worried by what microbiome testing shows. “At the moment, we just do not have enough knowledge to safely and accurately recommend any specific interventions (such as particular changes in diet, and/or particular probiotics) to cause beneficial long-term changes to the composition of the bacteria in the gut,” Guts UK said.
ZOE, however, told us that their data showed members “who follow ZOE advice, including gut booster foods, see an improvement in their gut microbiome score” — the mix of “good” and “bad” microbes. This conclusion is drawn from a ZOE internal survey of 450 members who followed their advice for more than 12 weeks, which also showed 70% have more energy, 85% improved their gut health and others saw improvements in their menopause symptoms. This is all heavily promoted across their website.
These results haven’t been published and they declined to share it with us when we asked, so we are unable to assess what was asked or how the sample was selected. Moreover, as Dr Guess points out, “the data has not been validated or peer-reviewed”. Neither was there a control group, which would be needed to know whether it was ZOEs’ specific advice, or simply better healthy eating, that was making a difference. When we relayed this to ZOE, they agreed this meant that we could not be sure these were “clinically meaningful results” and told us they were going to be publishing a randomised controlled trial in the future.
It is hard not to be blown away by the dizzying array of high-profile journals that feature on ZOE’s website: Nature Medicine, The BMJ, The Lancet, among others. It links to abstracts of papers published in peer-reviewed journals, which on the surface seem to provide proof of the science. At a closer look, however, not all is as it seems.
ZOE places the weight of their programmes’ claims on three main studies with the acronym PREDICT (Personalised Responses to Dietary Composition Trial). They were carried out with scientists from prestigious universities, including Stanford, Harvard and Kings’ College, London. They have not yet all been published.
The studies analyse the results of people making multiple readings about their body and food intake. This has produced a huge dataset: more than 4 million glucose readings so far, for example. These findings are then used to develop models to predict the impact of particular foods on our bodies — all valid areas of study. The researchers have searched for associations between different measurements, such as hunger and types of food eaten.
All of which sounds impressive. Until you realise that these were observational studies, which can only show associations between various data collected, and not causations, as Tim Spector himself has previously noted. For while measuring lots of variables to find patterns can be a useful research technique, it also means you increase the chances of finding coincidences — which puts you at risk of misinterpreting random variation as a meaningful difference. Poor practice in this type of study is known as “data dredging”.
Nor do they answer the key question: is there evidence that the individualised advice generated by the ZOE tests, which is the key market differentiator, does what it claims?
The answer: we don’t know. There have been no trials published exploring whether the “personalised” advice given by ZOE can result in better health outcomes compared with standard diet and lifestyle advice.
And the same holds for menopausal women, the key ZOE demographic targeted by Davina McCall. The first PREDICT study in 2022 compared data in pre, peri and postmenopausal women. According to ZOE: “Women who have been through menopause had, on average, higher blood pressure and blood sugar, a greater risk of developing cardiovascular disease in the next 10 years, worse sleep, and more body fat. Crucially, the team’s analysis found that diet and the bacterial species that were present in women’s guts were at least partially responsible for the changes.”
However, Dr David Nunan, Senior Research Fellow at the Centre for Evidence Based Medicine at the University of Oxford, is sceptical. “It’s a cross sectional [observational] study” he says, “which means that no claims of causation can be made.”
The design of the study, in other words, simply isn’t capable of telling whether any changes in the microbiome are a result of menopausal change. In fact, Dr Nunan says, “when the data for all microbiome species are compared, there is no statistical difference in pre and postmenopausal women. It’s only when you examine a subgroup of species that 8 out of 2,452 show a statistically significant difference.” Moreover, even when this happens, the study has not indicated what this means in real life. In other words, we do not know what impact this has for the ways that women metabolise food.
Yet despite these gaps, ZOE told us that they are “proud” to have recently completed their own randomised controlled trial “with a non-personalised control group versus the ZOE programme”. The full manuscript, they say, is “under preparation”.
According to its critics in the science community, however, even this trial has its design flaws. One group receives the whole ZOE package, the other receives a leaflet with US health advice. But the way it is set up, people will know exactly which group they are in, thereby rendering results less reliable. A fairer test to assess is its ZOE’s personalised algorithm that makes the difference would be for both groups to receive blood monitoring and stool analysis, and what they thought was personalised advice. The only difference should be that one group receives advice based on ZOE’s tests and algorithms, while the other receives advice that looks like it might be “personalised” but isn’t.
“You have to bear in mind when looking at these studies that they aren’t fully independent,” Professor Banerjee says. “They are marking their own homework. You need external validation and testing to reproduce the results.”
None of this is to say that paying close attention to nutrition is not important. ZOE is only one of hundreds of apps that measure our biometrics in this age of the quantified self. But while keeping abreast of our data might be interesting, are these promises of personalised advice based on sound medicine? And are they, as the website promises, scientifically proven to lead to real-world improvements?
Without this evidence, Zoe is simply collecting our data and then selling it back to us. And not only does this come at a financial cost — it could be making us less healthy.
Join the discussion
Join like minded readers that support our journalism by becoming a paid subscriber
To join the discussion in the comments, become a paid subscriber.
Join like minded readers that support our journalism, read unlimited articles and enjoy other subscriber-only benefits.
SubscribeMy personal experience with Zoe was that it powerfully reinforced what I already knew; a mainly but not exclusively plant based diet low in ultra processed food is not only far more healthy but leads to sustainable and controlled weight loss.
Any programme that encourages changing diet rather than going on a diet is far more useful to me and my patients than the weight loss wonder drugs we are seeing and being pressured to prescribe.
As a. GP I strongly disagree with the claim that we are unhappy with the increased workload from people doing Zoe or other continuous glucose monitoring. I am personally very pleased that my patients are taking responsibility for their own health and will actively support them, I only wish more would do so.
Good. I felt that the criticisms of the programme in the article lacked substance. Some of them seemed to be little more than ‘get off our turf’
If you can read that we’ll argued and evidence based assessment of Zoe as that, I’d suggest science and indeed critical thinking isn’t your strong point.
‘personal experience’ is not evidence, and indeed this comment does not include anything that can be described as evidence, rather being a case study in confirmation bias.
MF’s ‘personal experience’ does match up well with the bulk of the science, though, not least the massive ‘China Study’ and the work of Western A. Price in the 1930s (Nutrition and Physical Degeneration). You should encourage doctors who know about nutrition, because most don’t, as they’ve been dumbed down by pharma propaganda at medical school.
Hardly rocket science though.
On the other hand everybody knows weedy sickly people doing all the right stuff and boisterous 80 some-things still on beer, biscuits, curry and crisps.
I feel a venture capital pitch coming on
“ZOE” is just another nasty little tool invented in order to turn human beings into data streams for the corporate machine. Don’t be a part of the techno-fascism hellscape that is being built around us. Surveillance in every area of our lives. Repetitive, robotic voices penetrating our brains at “self check-outs” in supermarkets, ensuring we have limited human contact and progress towards sociopathy. Kids glued to phone screens instead of playing and discovering the natural world. Fight the over-digitisation of our lives wherever and whenever you can, and be more human!
Exactly. Not that long ago, people were obsessed with the gadget that calculated the number of steps they took in a day. How many are still doing that?
I recall a TV diet guru (with absolutely zero credentials or nutritional expertise) telling some gullible morning show hostess about her pineapple diet. The doctor who was on after the commercial break, a crusty, no-nonsense medical professional, scoffed at Pineapple Lady. “If you want to maintain your overall health and weight, get up off your rear and stop eating so d*mn much.”
People are fixated on themselves to the point of asininity.
Lots of people are still counting steps to be fair. I can’t see how that can cause any harm – exercise being generally a good thing.
To be fair to the article they were raising the idea that eating fruit and veg might be a good thing, only we don’t need to pay £300 a year to know that.
I agree. You can also sync your smart watch up to virtual walking apps like Conqueror which provide some added motivation, particularly if you choose one with street view.
I look at my steps thing on the phone, but just for a laugh really, a bit of daft conversation..I don’t think doing twic as many will make me twice as fitter or anything like that. And of course you don’t pay a wedge of money for the app.
As the person who twigged into faecal stool collecting might have said: “Where’s there’s muck, there’s brass”, and where there’s brass there’s always the danger of hype, hope and fashion charging in.
I quit Zoe today, by coincidence, because of the negative effects it was having on my health. What on earth did I think it could teach me, at the age of nearly 75, that I did not already know? My very wise late grandmother once said, you’re a fool or your own doctor by the time you’re 40. In other words, you have learned through half a lifetime’s experience how your body responds to foods, exercise, stress, the hormonal cycle. You know your weaknesses and basically what to do to keep yourself in reasonable shape. I do regular exercise, don’t smoke or drink and I eat a reasonably healthy diet. I’m neither fat nor thin. I’m on blood pressure meds, but otherwise this is probably as good as it gets for my age. Yet I too was seduced by the Zoe ads. One does fear what might lie ahead and grasp at anything that seems to offer a way of extending the fit part of old age as long as possible. So I donned the monitor, sent off the blood and poo, and tried to follow the regime. It has been a pretty unmitigated disaster for me, because my lifelong weakness is the most delicate and hypersensitive of digestive systems. So much so that a diet of yogurt and raisins in my 20s made me so ill they took out my appendix, before telling me it was just a gut problem. The Zoe test muffins duly made me vomit. The test results, telling me I process fat badly, my diet is poor and my gut microbiome lacking, had me in that typical “worried well” panic. Because although the test is personal and individual, the aftercare is not. You get a huge data dump with your results and deluge of very general recommendations with no way of prioritising what suits you best. In my panic, I cut out refined carbs and most animal fats. I forced myself to down gallons of kefir, eat yogurt and mountains of vegetables I really do not like. It’s about five weeks since I began the process. In that time I have been ill three times with stomach cramps, nausea and bloating that takes me right back to my miserable 20s. My sleep was all over the place. I’d have strange energy bursts at 2am that kept me up for hours. Today has pretty much been spent on the loo so far. And when I wasn’t shopping for new foods or trying to make them palatable, I was watching videos and filling in surveys. It dawned on me that I was working so hard on trying to prolong my life that I had stopped actually living it. So back to what I have worked out for myself over a lifetime living in my particular body. I can’t digest fermented foods or many vegetables. So what? Jordan Petersen’s daughter Michaela eats nothing but red meat and looks fantastic. It takes all sorts and there may be no generally applicable rules. I’ll probably keep the refined carbs and processed foods to a minimum. But once I recover from last night’s vegetable soup with brown rice with a side of avocado, salad and beans, I am going back to my old way of life. Perhaps the lesson learned was worth the £299 I spent.
I’ve been using Zoe for about 4 months. The initial analysis was interesting (I’m good at processing fat, but bad at processing carbohydrates) but it seems to then be scared of following through on the logical advice from this data – it’s ok for me to eat fatty things. Ultimately, the advice and meal plans are just pretty basic common sense – more vegetables, less sugar, less meat. Like this article, I’m concerned that a lot of the things they declare as facts, particularly around gut health, seem to be driven by observational studies and the bias of the lead scientists at Zoe. They don’t seem to be able to convincingly back up their boldest assertions.
Ultimately, I don’t think this does any great harm to anyone, but as the article says, it’s generally targeting wealthier, healthier people (like me), not people who would really benefit from a better diet. And it’s expensive. I won’t be renewing the subscription, but I don’t regret trying it out.
Exactly my experience. Furthermore, after I’d started, I was “diagnosed” as prediabetic and therefore precluded from the study/plan. In my opinion, I was an ideal candidate for testing its alleged benefits, even though I also feel (in the context of diabetes) that there is something of an obsession with blood glucose levels in isolation, rather than viewing them as a proxy marker for insulin sensitivity (I’m trying to avoid – perhaps unsuccessfully – to be too pedantic: sometimes different phenomena amount to the same thing – and hence practical consequence – even though they are not the same thing). With more specific respect to ZOE, there is a long history of the unreliability of self-report/anecdotal/observational studies, ESPECIALLY in the areas of dietary control/ weight loss/obesity: basically, people delude themselves. I’m not suggesting this necessarily applies to gut health studies, but the transition from gut flora science to the epidemiology of Covid, and back again, did bother me somewhat! As far as I know (which is not as far as I’d like) the initial detailed PREDICT analysis is based on sound empirical science, although the degree of detail did seem at first spurious. And I’m not sure to what extent it encompasses what must be huge individual responses; an awful lot depends on the sophisticated statistical expertise of the likes of Tim Spector (who incidentally has the knack of always coming across as balanced, unruffled, while regularly critical of government policy). As for ultimate dietary recommendations, which have, not just through ZOE, developed into holy writ, we have low-carbs – absolute minimal refined -, avoidance of highly processed, plant-based and fresh greens but not necessarily vegan, vigilant about too much red meat, awareness of the “low-fat” deception, and – a ZOE speciality – plenty of fermented foods. All of which is not cheap – unless you eat a lot less than I do!
Most of these applications seem to follow a similar pattern.
My own experience (not ZOE) tends to be that during periods of high activity over a good summer – lots of cycling and hiking – I spend far too much time looking at stats in Garmin, Strava and tracking diet in MyFitnessPal. The weight falls off, which is hardly surprising when racking up 100+miles of cycling and 20+miles of walking in a week. The improved (but not great) diet tends to be a result of craving a wider range of food.
Come winter, when this stuff would be more useful as the exercise is probably 20% that of the summer and the diet probably twice as bad, I just don’t bother with it. Although I’m not necessarily “typical”, I would expect similar patterns of positive reinforcement in many people.
Makes sense- a standout virtue on social media comments pages these days.
Just listen to the free podcasts, they are interesting, informative and varied. The speakers inspire trust – or not. Just use your judgement. I don’t feel any need to sign up for any elaborate, faddy pseudoscience designed for the “worried well” with money to burn
Just stop eating garbage.
I was a vociferous fan of Tim Spector and his Covid app as I believed he seemed to want to help society, but at no great monetary gain to himself. It was a brilliant way to collect population data on Covid prevalence, and worked alongside the randomly selected ONS (office of national statistics) data.
However, my high opinion of him began to wane when he was angry that the Government refused to continue his funding as the pandemic resolved. One cannot blame him for monetising the Zoe app, but one cannot get away from the adverts, and IMHO it is a luxury buy in « developed » countries for anyone one that has time to read a 30 page document about their gut, and the tendency must be to become somewhat obsessed about your blood glucose.
It’s just the latest fad. It will dwindle over time, and in the meantime I will increase my vegetable, fruit and grains consumption and take the dog for a good walk, whilst saving myself £300.
Have you pitched your last paragraph at any Venture Capital groups…. could be an app in it?
My wife has been using Zoe for quite some time now.
She feels much better and healthier.
I agree with Mathew Fox on this. In my view her benefit has come from the regime of monitoring foods, knowing a treat needs paying for the next day. Finding out which foods to avoid except on an occasional basis. For my wife Zoe has been a game changer.
Her metabolism has always seemed a bit strange and after years of “dieting” the Zoe regime seems to have had many real benefits.
As a general rule of thumb, it’s best to take panaceas with a large grain of salt. Coarse, naturally sourced sea salt, of course. From an online company based in San Francisco. Salt from the grocery store simply won’t do.
Seriously, though, good job addressing the specific points that challenge the credibility of this latest fad. The science just isn’t there. But your point about this actually harming people’s health is what really sticks in my mind. Take a healthy person, make them anxious about their health, give them a product that promises to help them take charge of their health, and they wind up even more anxious than they were in the first place because all the product does is give them excessive, raw data about one aspect of their body’s functioning in real time.
Ain’t capitalism grand?
I joined ZOE in June and I’m reaping the benefits!
Zoe has steered me towards eating more vegetables, pulses and beans. My sugar craving are completely gone, in fact I can’t remember the last time I had a bar of chocolate or a pudding (and I used to eat a lot of chocs). The app makes you choose better options and as a result I feel less bloated.
I was hoping lose a few pound, which hasn’t happened, my weight remains the same. I have a cholesterol check up coming up, this will be the ultimate test. I would be very disappointed if my bad Cholesterol hasn’t come down significantly.
We all choose how to spend our money and my health is important to me.
So you needed to pay hundreds of pounds to implement advice any doctor in the NHS would have given you for free?
It’s a funny old world.
What is not sufficiently highlighted in this article is that we are all individuals and that ‘normal’ does not quite exist. A living being is a system that interacts with his environment which is guided by cognition and tries to survive by managing a dynamic balance. Health is when this mechanism has lots of flexibility. The less flexibility it has the less well it responds and starts making/repeating mistakes.
The word cognition is the/a difficult concept : I borrowed it from Capra (system view of life).
I essence we each tend to respond in slight different ways: that is why having ‘one solution for all’ does not work. Further we are such complex systems that the simplified health message and even a accumulation of data only cover an aspect of what health is. In the end it is a narrative of which large parts are shared by many but of which smaller parts are only shared by some or maybe only even one person.
Therefore: Zoe will work for some and not for others, however much research will be done. The narrative it uses will resonate with some and not others: it is a narrative like so many other health narratives that work for the one and not the other.
Also, there are more and more voices agreeing that looking for good medical evidence though Double blind trials and the like works only if you examine narrow subjects in medicine. You cannot examine health through such ‘scientific’ means.
But more money is to be made with narrow health narratives: there are always enough people with whom they resonate, which attracts investors..
Of course health can be examined through RCTs, that’s the field that primarily does them.
If it’s all about narratives though, why is it being marketed as science?
I propose you read Iain Mcgilchrist: The matter with things. Or listen to a few of his podcast.
And no, you cannot examine health through RCT’s unless your definition of health is to have as many biochemical measurements as possible within what is considered by the lab that tests you to be within normal. In that case you put no value on your personal experience… this is how machines run …..
Modern. Life. Is. Rubbish.
I signed up for this because I was interested in gut health. The initial £600 I think is worth it – there are blood and gut fauna tests, including responses to a standardised sugar load under different conditiions (they send you a pack of buns for this) and two weeks of a blood sugar monitor.
All this undoubtedly gives useful feedback on personalised responses to particular foods. Going into the thing knowing a bit about what I wanted out of it was useful, and it confirmed my suspicion that my diet gives me good gut health, but that my blood sugar response needs work. Two weeks on a blood sugar monitor, logging every meal, is quite enough though – I wouldn’t want to live like that, logging everything and watching the feedback in real time.
There were some major down-sides. The ‘nutritionists’ supposedly on-call were pretty useless spouting, generic stuff you could read (and probably have read) in a daily newspaper’s magazine section or on one of those interchangeable lowbrow youtube ‘health’ channels. No attempt to engage with individuals’ specific needs, such as discussing possible drug interactions.
Then there was the app itself – which was ok as far as it goes (for those sorts of people who love filling in on-line forms). But it **is** only an app – no web site available for older people who unlike Millenials, whose thumbs have evolved in a super-human direction, and who prefer interacting on-screen. And really, it’s no different to other lifestyle apps that nag you to do healthy stuff, except it starts with a bit more info than they do.
So I jacked it in after a month. I couln’t see the point of paying £30 a month for useless cheerleading off-stage, and an app that irritated me and that I couldn’t be bothered filling in beyond the end of the blood sugar report.
A big problem though is that you can’t take much away afterwards without taking a million screen shots – no way to get out of it a proper PDF report that you could, for example, then take to a GP or a proper nutritionist.
So personally I think the whole thing is doomed to financial failure. The initial stuff on blood sugar, protein response, gut fauna health and so on is undoubtedly useful, but it’s not cheap to mail out buns, blood sensors, stool packs, run the tests and analysis, and generate reports. It’s clear the business model relies on milking the £30 a month crowd who persevere with it beyond the reporting stage.
I know a few people who have signed up for it but no-one at all who have continued after a month Speaks volumes.
(edited for spacing)
As someone currently in the testing phase on Zoe, as one of the worried well, so far it’s not doing much apart from reinforcing that I can’t eat carbs without protein or fat otherwise I get a massive sugar spike, attendant headache and then the need to eat properly. I feel slightly silly for signing up to something so expensive. The lack of knowledge about glucose spikes and dips being natural after eating, and how on earth I flatten the curves is frustrating but I will continue to the end of the “trial”. At my age of 47, I know what I should eat, what works for me, but I still don’t stick to it rigidly. Life is too short to not have a Pret brownie occasionally, and in fact feel no worse than a revolting bowl of porridge made with nothing but water!
I haven’t used the Zoe food app but it was obvious that their Covid app which I did use suffered from having a middle class, female bias in the sample size. The algorithms they used to show the number of cases had to be revised a number of times when the results differed significantly from reality.
Given that Zoe were then looking for only one variable, Covid or not, suggesting they can now make accurate predictions based on a smaller, probably equally biased sample size, on a subject with multiple (possibly many thousands) of variables does seem unlikely.
Very interesting. I’m a great fan of all this stuff in principle, and I do think that much of the acclaim around personalised health metrics and trackers is well deserved.
However it does rather looks as if it’s not really ready yet. I know from my own fitbit, which is very useful for activity tracking, that the sleep tracking part of it gives little value, being based as it is on proxy measurements that cannot possibly hope to be an accurate way to measure what it says it’s measuring.
But it will improve, for sure, and if the industry generally adds even only small amounts to people’s healthspans, it will be a good thing.
If you use a FitBit, you might be interested in checking it out for accuracy against more sophisticated (laboratory standard) devices. “The Quantified Scientist” on YouTube is a post-doctoral researcher in biological data analysis who has, to date, bought and tested over (?) 70 such devices — including a variety of smartwatches — mainly on himself, plus a female accomplice.
https://youtube.com/@TheQuantifiedScientist?si=xyDKVSg5RgfEET3S
I was interested in these mainly as a lifelong insomniac, after following a certain Dr Dale Bredesen whose ReCODE protocol (diet, exercise, sleep etc) has now been shown to halt and even reverse the progress of Alzheimer’s. Aged 81 with odd and infuriating memory lapses, sleep is a major concern.
According to a medical research paper posted by Bredesen on his FB page, we should all be getting an hour a night of slow wave sleep (“Delta”, at 0 – 4Hz electrical oscillations in the brain). “Polysomnography” measures this in sleep clinics, requiring a scalp covered in electrodes. But there are cheaper consumer devices (at $1000 to $2000) which are almost as good, which “The Quantified Scientist” uses to rate eg smartwatches against.
I discovered from his YouTube videos that Amazon’s cheap “Amazfit” wrist worn devices are so bad they are a complete waste of money. But even the high-end and expensive Galaxy Watch 5 Pro is not rated very highly, and is fairly abysmal in practice. I ditched it after it told me I had slept for 11 hours 15 mins after turning the lights out at 2am and waking at 7am.
However the QS has also quantified (in percentages) how far each device mistakes different sleep stages for each other: and I find the most common error for both the Watch 5 Pro and the Apple Watch 8 (as one of the best for sleep tracking) I am now trying, is confusing wakefulness with shallow sleep.
The main info I’m getting is that I very seldom exceed — or even reach — 30 mins of Delta sleep, during which the body regenerates cells etc.
But knowing what I do about the inaccuracies of all these devices, I’m hardly going to be hooked on its results. I find the Apple Watch 8 most useful for zapping my wrist if I haven’t moved for 50 minutes: telling me to Stand up! Move! for at least a minute.
The QS also downloads raw data to look at the algorithms which translates the measurements, and finds that some of them can be rewritten for more accurate results — ie that better match results from his more sophisticated devices.
I’ve been following Zoe since the beginning of the pandemic, but am not yet convinced to shell out on its faecal testing and dietary advice re gut microbiome: partly as I’ve been vegan for nearly 50 years, and already eat most of the stuff they advise in Zoe YouTube talks. Beyond that, Will Bulsiewicz gives good enough advice as a gastroenterologist.
This interesting and useful UnHerd article persuades me not to change my mind.
Very interesting, thank you.
I have the Zoe app and find the free podcasts very informative and sometimes eye-opening. I haven’t gone on the Zoe programme because I thought I could get much the same benefits by modifying my diet. There is no hard sell by Zoe. I think it has improved my health and wellbeing.
I assume Zoe is ruffling a few feathers amongst producers of ultra processed foods in particular so it wouldn’t surprise me in the slightest for ‘experts’ funded by these companies to be gunning for them and trying to attack their credibility. Is there a bit of that here?
It is possible to have two things, which, tho diametrically opposed, are both utter con jobs.
By chance this article has arrived in time to save me. I was being sacked in to the system. I’ve paid for a 4 month subscription but yet to receive the test kits etc… to start. I will now be viewing it all with a more sceptical approach. Having said that, I feel I have benefited from listening to the Zoe podcasts and will continue to use them as a distraction when driving. I will certainly not be spending any more money on this. It is sad that, what seems to have started with the best of intentions, has been hijacked by those who see it as a cash cow.
??? A lot of the time, bacterial stool tests tell you sod-all about what’s even in your stool, let alone what’s going on higher up in the digestive tract
Just come across this article and the comments both of which were excellent. As a user of the ZOE website – I do think their short videos on nutrition and exercise are excellent by far and away the best on the market e.g. recent videos on the best way to build muscle after the age of 50. Agree with the comments of the authors on the microbiome and the dietary advice. However, I think they could have gone into more depth on Continuous Glucose monitoring for healthy people. There is a very good section on this in Peter Attia’s book Outlive pages 323 to 333. He recommends that healthy people should only use if for a month at the most this being the insight phase leading to the second phase behaviour modification.
I’ll admit I didn’t finish the article, because I can spot a scam. What just slayed me was the photo of Tim Spector (center) and his “staff.” The women on the left side of Spector are the exact same women as on his right side. So he has expanded his mission to cloning people?
Twins study
To be fair, Spectre has a very big longitudinal identical twin study. Those doctor twins who appear in BBC popular health programmes were part of it.
One went to live in USA and became obese…. generated a book looking at diet v genetics. However it all comes back to common sense eating as always.
I followed Prof Spector during covid, mainly because I was desperate to see the back of lockdowns and get my life and freedoms back. I felt Spector took a reasonable view about the number of cases and how covid would soon become more prevalent and much less alarming and he expected we would soon be able to go about our daily lives as normal. Suddenly however his narrative changed and he was all for continuing lockdown, masks and the vaccine. At the time I felt he’d been pressured by the powers that be and that it probably came down to funding. I went off him after that and lost interest in his microbiome stuff too. I thought he was unreliable. Seems I was right to think that.
X
A helpful perspective – thank you to Deborah and Margaret for their work on this. As nutrition scientist Dr. Alan O’Flanagan put it when referring to these new obsession with personalised nutrition, “the enthusiasm outpaces the evidence”. Is anyone aware of other critiques/concerns of Tim Spector and/or Zoe out there? I’m only aware of one other critique from Layne Norton (a US based nutrition scientist) who took umbrage with several claims/statements that Spector made in a podcast appearance on “The Diary of a CEO”. Here is the link: https://www.youtube.com/watch?v=OVu7YyMXGBM&t=11s