If there’s one thing the long history of nutrition fads teaches us it’s that, whether they’re prescribed for weight loss or healthy living, blueprint diets have their drawbacks.
With the vinegar diet, for example, popularized in the 1820s by the English poet and legendary narcissist Lord Byron — essentially drinking cups of neat vinegar with every meal — the main drawback was getting through the day with a mouth like a chemical toilet (plus copious bouts of vomiting and diarrhea). Later, with the tapeworm diet, reputed to have emerged among corset-conscious Victorians in the late 19th century, it was introducing a damaging parasite into your body (supposedly via a pill containing tapeworm larvae) that made you feel ill, then proved very difficult to get rid of. With the chewing diet, prescribed by America’s ‘Great Masticator,’ Horace Fletcher, at around the same time, it was forcing yourself to chew each mouthful of food at least 100 times until it slushified, then spitting out any remaining solids like a giant carcass-draining spider — not to mention only producing feces once every two weeks (though on the plus side, they smelled like “warm biscuits,” according to Fletcher). With the Atkins mainly-meat-and-protein diet, wildly popular in the early 2000s, it’s bad breath and (as a recent study has suggested) long-term weight gain.
The pre-eminent downside of today’s crop of faddish formulas, meanwhile — from runny juice diets to the ancient nuttiness of Paleo to the low-carb crash of keto diets, and even of the general healthy-eating regimes widely recommended by doctors and mainstream nutritionists — is buzzkill. The thought of chomping down whole grains and fruits, rather than the tasty, doughy, sugary carbohydrate-y mush-food we grew up on, can sometimes feel like a lot of hard work for a lifetime of bland.
But as nutritional science is increasingly acknowledging, it might not be the gastronomic gloom of dieting regimes that makes them so difficult for so many of us to stick to. It might be, literally, our gut instinct that’s putting us off. “We now know that a ‘one-size-fits-all’ approach is too simplistic,” says Sarah Berry, associate professor of nutritional sciences at King’s College London, whose research is helping to foster a growing appreciation in the scientific community of just how variable our “ideal” food intakes really are. Within doctors’ “broad healthy guidelines,” she says, “we will all respond differently to the same food.”
Lately this realization has combined with the 21st century’s relentless commercialization of personal data to launch what might well turn out to be the food fad to end all food fads. Personalized diets are being marketed by tech-driven nutrition services that have jumped on the revolution in commercial genome analysis and are hoping to ride it right back down our throats. First-wave companies are offering eating plans that are fine-tuned to their customers’ individual dietary sensitivities and metabolic quirks, as revealed in their DNA, which some of these services measure alongside a number of factors such as their age, activity levels, food preferences and a series of metabolic markers found in the blood.
It all sounds reassuringly scientific — and with the promise of plans that will take into account things we actually like to eat, appealingly non-diety. But it also has that frightening ring of tech utopia that wants to claim life’s toughest problems can all be solved with a well-designed, clean-lined app. So should we swallow it?
“Yes, I think there’s enough there at the moment for them to say, ‘It’s worth personalizing diets,’” says Jennie Brand-Miller, professor of human nutrition at the University of Sydney and director of the Glycemic Index Foundation. “It’s at an early stage,” she cautions, “but there’s some truth in the idea of first [to market] gets the recognition, and there’s a great deal of opportunity there.”
You Say, “Tomato,” I Say, “No Thank You, I Have a Mild Intolerance to Lycopene”
The early bird might get the tapeworm, but the big hiccup with personalized plans is that there’s a vast amount that isn’t yet known about how our food intake affects our bodies. Exactly how different are we from each other? “That’s where the research is heading,” says Brand-Miller, but “at the moment we don’t know the answer.” Genetic divergence with our species is certainly one part of the story, and she illustrates how wide these gaps between our guts can be with a couple of well-studied examples.
The first is the fact that only some of us are able to fully digest lactose, the sugar in milk, beyond childhood. Far from being the rare, debilitating disorder we hear people bragging about to baristas, lactose intolerance, or at least a mild form of it, is our natural evolutionary state and the majority of the world’s adult population, like other mammals, greatly reduce the amount of lactase (the enzyme that breaks down lactose) they produce once it’s no longer useful for weaning. (For some people, of course, it’s a genuine problem, and they really shouldn’t be served lattes.)
One knock-on effect of this split in the global population along milk-digestion lines, explains Brand-Miller, is that our lower intestines become a further source of deviation — specifically in its impact on the bacterial colonies known collectively as your intestinal “microbiome,” understood to be a vital component in how we process food. For the 65 or so percent of us who aren’t producing much lactase as grown-ups, she says, “your microbiome are getting a source of food — the lactose that’s undigested. And since only certain bugs can digest lactose, the make-up of the microbiome will be different.”
Another illustration of how the food we eat has forced genetic differences in human beings is in the ways our bodies deal with starch, a carbohydrate that has to be cooked in order for us to glean any calories from it. The enzyme that breaks down starch is called amylase (the main ingredient in saliva), and per Brand-Miller, “Humans are unique in the case of all the primates: We have varying numbers of copies of the amylase gene. Some of us have got 20 copies, and some of us have only got two copies. That means that something during the course of evolution encouraged the duplication of this gene.”
Having more copies of the amylase-producing gene “is a really interesting phenomenon,” she says, “and we’ve done some research recently showing it does in fact mean that starch gets in faster to the blood.” Other studies have suggested that people with lower copy numbers for this gene could be at higher risk of developing insulin resistance — which can lead to diabetes — and some researchers have also found a link to obesity (though others haven’t found the same connection).
Complicating the picture vastly, though, is that in other aspects of metabolism, our DNA seems to be less of a determining factor than you might assume. Blood glucose responses, for example — or how rapidly and intensively our blood sugar spikes after we’ve eaten carbs — have important effects over time on a range of outcomes for our bodies. “The higher they are, the more likely we are to develop diabetes,” explains Brand-Miller, adding that they also play a role in cardiovascular disease, “and maybe even in cognitive function and dementia.”
How each of our bodies regulates our blood sugar might seem like a shoo-in for a genetically inherited trait, and she says, genetic factors like lactase persistence and the number of amylase-gene copies you’re packing in your genome do indeed affect glucose concentrations in the blood. “But the major differences that explain more of the variability are environmental things, like your weight, your age, how much muscle mass you’ve got, how much sleep you’ve had the night before the test and drugs as well.”
You Say, “Potato,” I Say, “Major Source of Dietary Starch”
This downplaying of DNA as the dominant source of our nutritional differences is a pattern that’s also emerged in a pioneering project that’s taking place in the U.K. right now, in collaboration with scientists from institutions in the U.S., Sweden and Italy. The enterprise, called Zoe, bills itself as “the world’s largest scientific nutrition project,” and it’s based on the results of a series of studies that are currently underway, led by the Department of Twin Research at King’s College London. Berry, one of the scientists involved in the research, says that so far, “The most surprising outcome from our study is just how large the variability in responses is between even healthy people, and how little of this variability is due to our DNA.”
The ultimate aim for Zoe is to create a home food-response test and an app that will join the other personalized diet products out there vying for consumers’ attention — but one that’s based on a more complete scientific understanding of the complex interactions that shape our metabolisms. In its first phase of study (the results of which were made public in June), the researchers monitored food responses — including blood sugar, insulin, fat levels and other blood markers — in 1,100 adults in the U.K. and the U.S., 60 percent of whom were twins. “By studying identical twins we were able to perform the ultimate genetic test,” says Berry. “We found that identical twins who share all their genes and most of their environment had very different responses to identical foods.”
This unexpectedly large variation, she says, was only partly determined by genetic factors, “with less than 50 percent of the glucose response, 30 percent for the insulin response and 20 percent for the blood fat response being explained by genetics.”
Aside from these three classic markers of food response, the team has looked in detail at a range of other factors, both biological — such as the microbiome — and situational, such as when the subjects ate their meals and what was actually in them. “I was also really surprised by how the nutrient composition of the food itself didn’t play a bigger role in determining responses,” says Berry, who explains that differing amounts of fat, protein and carbs in meals only accounted for less than 40 percent of the variation in people’s responses to them.
Another finding she and her colleagues hadn’t bargained for was the influence the contexts surrounding our meals seem to have: “We also found that the time of day, timing of exercise, how much sleep people had and the composition of meals earlier in the day played a role in determining responses to food.”
These results clearly show that as individual eating machines, we’re much more hand-crafted bespoke models than production-line duplicates, and seem to suggest that personalized diets are likely to be the only form of planned eating that’s a safe bet to deliver the physical results we want. But they also show the extent to which “DNA is just one of the factors that determines our responses to foods,” and, says Berry, “Our results highlight the limitations of the other companies offering tailored dietary advice based on only a snapshot of these factors.”
According to Berry, a detailed analysis of the microbiome is another glaring omission from the personalized eating solutions that are currently available. “We’re starting to understand that the trillions of bacteria living in our gut play a key role in determining how our body metabolizes food and subsequent health effects,” says Berry. In the research for Zoe, it’s emerged that, “Our gut bacteria differs hugely between individuals, and even identical twins only share 37 percent of their gut microbes.”
The twin study has also revealed that our microbial freeloading friends play an important role not just in shaping our glucose responses to food — those spikes in blood sugar that can have such drastic long-term effects — but also our blood fat and insulin responses. “This is a really novel and exciting finding,” says Berry, “and has really advanced our understanding of the importance of the gut microbiome.”
Universally Prescribed Diets? Let’s Call the Whole Thing Off
The whole question of how deeply personal a personalized diet needs to get in order to be effective is complicated even further by another kind of culture. As Brand-Miller points out, there’s little point in carefully tailoring a nutrition plan for someone if they can’t wear it. “Fifty years ago we knew that diets differed around the world and that people tend to like the diets that they grew up on,” she says. “So a Mediterranean diet with lots of olive oil and other things will suit some people and not others. Dieticians know that you’d be crazy to recommend a Mediterranean diet to someone of Asian origin; it’d be much better to devise a low-G.I. diet that’s suitable for someone from India, Vietnam or China, and another form of low-G.I. diet that has some appeal to someone who likes a Mediterranean diet.” All of which has more to do with psychology, culture and personal preference than internal metabolism, but she says, “still comes under the term ‘personalized.’”
A ‘low-G.I. diet,’ by the way, refers to opting for foods that are low on the Glycemic Index, a system of rating the speeds at which carb-heavy foods are digested (the slower the better) and their potency in stimulating our blood glucose response. It’s a scheme that Brand-Miller has done much to elaborate and promote in her career thus far: Focusing on food that tends to provoke a slower build-up of blood sugar after meals (such as whole grains, green vegetables, and fruits) is widely thought to provide a range of health benefits, especially for those with diabetes or who are most at risk of developing it. But isn’t this regime, along with other sets of healthy-eating guidelines widely promoted by governments and public-health bodies, just one more outdated dietary prescription that’s being undermined by an increasing taste for personalized nutrition?
Both Brand-Miller and Berry think it’s not necessarily a case of one or the other. “There’s still a place for dietary guidelines in parallel with bespoke dietary advice,” says Berry. “We should all be aiming to consume a variety of foods, reduce our intake of animal-based fats (except dairy), increase our intake of whole grain and fiber-rich foods and increase the amount of vegetables and fruit in our diet.”
According to Berry, this is valid advice that can be applied as a framework for healthy food species-wide, but within these parameters, there’s ample room for variation and refinement. “By providing bespoke advice, we can provide individuals with the knowledge of how to eat best for their individual metabolic responses.”
Brand-Miller, meanwhile, is all in favor of the new market for personalized nutrition plans, even in their current premature and exploratory form. “I’m someone who wants to see people make positive changes to their lifestyle in order to improve quality of life and to reduce the health-care budget,” she says, envisioning tech-led bespoke dieting products as one more way of encouraging people to take an interest in their health. “My feeling is that you need to press the right buttons to get behavior to change, and personalized nutrition might be one of those buttons.”
Illustrating how pressing the buttons — or tapping the screen — on a highly customized dieting app might usefully dovetail with the kind of blanket low-G.I. recommendations publicized by her own foundation, she offers an example: Let’s say a personalized diet company has shown “that you’ve got a predisposition to diabetes; you’ve got 300 of the 400 gene mutations associated with getting Type-2 diabetes — it would then be important for you to take note of the most beneficial carbohydrate foods.” Good point.
So it seems as though that the data-crunching power of technology is now at the point where it can cope with the dizzying array of internal processes and external influences that contour human nutrition. The next step is understanding all those interactions in the sort of fine-grain resolution that’s needed for firm, medical-grade advice. Insights from studies like the Zoe project, says Berry, “combined with machine-learning expertise is enabling us to build algorithms to predict an individual’s response to food.”
But even if a fully comprehensive personalized readout on what we should be feeding ourselves is some way off yet, the steps taken so far are already changing the way we’re appraising what’s on each other’s plates. “Nutrition research is seeing a big shift away from looking at nutrient-only advice and recognizing the importance of food-based and dietary-pattern advice. We’re also recognizing the oversimplification of looking at population averages when determining the health effects of foods.”
The Zoe project’s first round of findings, meanwhile, is an early indication of “the magnitude of the variability between apparently healthy people, how little of this is to do with the composition of the meal itself, and most importantly,” says Berry, “how much isn’t genetic, and therefore, modifiable.” For her, that makes the future of personalized nutrition “really exciting, for scientists and individuals. We really do have the power to modify our responses to food and our long-term health.”
And what of the old-style programmed diets? Whether they’re founded in vinegar vanity, weight-loss wishful thinking or the best available statistical evidence for healthy populations, if they’re not flexible enough to accommodate the whole menagerie of metabolisms that are out there, they might not survive too long into the age of personalized nutrition planning. Just like the promises they so rarely seem to deliver on, they will carry little weight.