Low-carb diets will shorten your life? Fat chance!

By Marika Sboros

Well, here’s a page turner for the diet books. A new study in The Lancet Public Health claims that low-carb diets could be killers.

The researchers claim that the diets increase your risk of mortality (premature death) by shortening your lifespan. In other words, the researchers argue that low-carb diets are life-threatening.

They also claim the same for high-carb diets. But don’t mistake that for any kind of anti-carb stance. On the contrary. The researchers claim that “moderate” carbohydrate consumption (50-55% of the diet) is the way to go.

They say that moderate carbs from plant foods up your chances further of avoiding a premature end. They speak of  “controversy” around low-carb diets. Therefore, they claim that low-carb diets containing animal foods are even more life-threatening than those with plant-based foods.

Overall, they seem to have a serious beef (pun intended), with low-carb diets.

It’s putting it mildly to say that leading doctors and scientists globally dispute those claims. And not just because this is yet another observational study.

Prof Richard Feinman

That is a big problem because, by definition, observational studies are only associational. Thus, they cannot prove causation. But there are other reasons that this study has left many critics hot under their scientific collars.

The kindest they’ve said so far is that we should simply disregard the study and concentrate on “more important” matters. That’s what Dr Luis Correia, director of the Centre for Evidence-Based Medicine in Brazil, thinks of it.

Other experts say that we cannot ignore it because it bears such a strong stamp of approval from Harvard.

Ugly moment in medical history

US biochemistry professor Richard Feinman says that in a field “not known for thoughtfulness”, The Lancet study ranks as “the most intemperate and irresponsible in recent memory”. In an email to me, Feinman calls it “a very ugly moment in the history of medicine”.

He says a big danger in this study. It will encourage doctors to withhold recommendation for low-carb diets to patients with diabetes despite evidence that these diets are “often a cure”.

Click here to read: ‘Pure’ proof fats don’t kill, dietary guidelines wrong? 


British consultant cardiologist Dr Aseem Malhotra is similarly scathing. On BBC World News this week, he called the study “fatally flawed” and a “miscarriage of science”.

Reviewing all the up-to-date evidence shows that low-carb are not dangerous, Malhotra said. And anyone claiming that we should all eat carbs moderately is giving “bogus dietary advice”.

Thus, the real meat of objections to the study includes cherry-picking data – ignoring growing robust evidence to the contrary. That feeds into other major objections that the study lacks scientific integrity at best. At worst, that it could constitute scientific misconduct.

More questions than answers

So, who are the researchers who would have you believe that low-carb diets really are dangerous? On what evidence do they claim that a moderate, plant-based carb intake is optimum for longevity?

Prof Scott Solomon

The study is titled: Dietary carbohydrate intake and mortality, a prospective cohort and meta-analysis. Of the 10 researchers, eight are from Harvard Medical School. Corresponding author is Dr Scott Solomon, Harvard professor of medicine and director of non-invasive cardiology at Brigham and Women’s Hospital in Boston, Massachusetts. The hospital is an affiliate of Harvard Medical School. The other two researchers are from the University of Minnesota.

Among Solomon’s co-authors is another Harvard professor of medicine as well as nutrition and epidemiology (now emeritus) Walter Willett. Willett is as well-known for his advocacy of plant-based diets and antipathy to low-carb diets as Malhotra is for his support for animal foods and low-carb diets.

The Lancet researchers studied 15,428 adults aged between 45–64, in four US communities during a median follow-up of 25 years. Their aim: to investigate the association between carbohydrate intake and mortality.

Data came from a dietary questionnaire that participants completed at enrolment in the 1985 Atherosclerosis Risk in Communities (ARIC) study. That questionnaire has raised more questions than answers.

Meat-heavy diets a killer?

In giving background, the researchers say that low-carb diets are a “popular weight-loss strategy”. They also say that the long-term effect of carbohydrate restriction on mortality “is controversial”.

They made it clear that they were only making observations. The researchers said that both high and low percentages of carbohydrate diets were associated with increased mortality. They claimed to have observed “minimal risk” at 50–55% carbohydrate intake.

Click here to read: Sat fat causes heart disease? PURE bollocks! 


They also claimed that low-carb dietary patterns favouring animal-derived protein and fat sources, from sources such as lamb, beef, pork, and chicken, were associated with higher mortality”. And that those favouring plant-derived protein and fat intake, from sources such as vegetables, nuts, peanut butter and whole-grain breads were associated with lower mortality.

That led to predictably fevered media headlines claiming that “meat-heavy, low-carb diets can kill”. And that cutting back on bread and pasta could shorten people’s lives by up to four years.

A UK report in The Guardian said that it could look like years of scientific study finally proving “what common sense already knew: everything in moderation”. Of course, The Lancet study proves nothing of the sort. If anything at all, it proves that common sense isn’t common.

The same report made a detour into what Aristotle would have said about moderation. The author sensibly (to my mind) added the caveat that moderation is not that simple. And even common sense recognises that if “everything in moderation” is true then moderation “also has its limits”.

Conflicts of interest

If Oscar Wilde were around, he would have heartily endorsed that sentiment. Wilde once memorably advised: “Everything in moderation, including moderation.”

Robust evidence for that is available. Britain’s NHS (National Health Service) GP Dr David Unwin has been using low-carb diets successfully for some years now to treat diabetic patients. He says that “telling diabetics to eat carbs moderately just leaves them moderately poisoned.”

Dr Aseem Malhotra

In media interviews, Malhotra says that anyone who tells people to eat everything in moderation is  “either ignorant or paid by the food industry”.

“Seek their conflicts and you’ll find them,”  he says.

The most effective approach for managing type 2 diabetes is cutting sugar and starch, Malhotra says. And when it comes to the evidence, a  systematic review of randomised trials reveals it: “Low-carb is best for blood glucose and cardiovascular risk factors in both short and long term.”

On Twitter, US psychiatrist and low-carb advocate Dr Georgia Ede has called The Lancet study “epidemi-illogical”. Studies of this type show a “complete disregard for mechanistic plausibility”, she says. That’s often to the point of “flying in the face of evolution, biochemistry, physiology, botany and common sense”.

She also noted that the researchers did not include or reference the questionnaire used in the ARIC study as a source of their data.

Weak science, strong personalities

It was left to that bulldog investigative reporter, US science writer Nina Teicholz, to find and make the questionnaire available. Teicholz is author of The Big Fat Surprise, Why Butter, Meat, and Cheese Belong in a Healthy Diet. It’s a seminal work that many say changed the face of nutrition science and evidence forever – and for good.

Nina Teicholz

Former editor of The BMJ, Dr Richard Smith said that Teicholz did “a remarkable job in analysing the weak science, strong personalities and vested interests in political expediency” that have characterised nutrition science.

Critics say those elements are at work in The Lancet study.

Teicholz has said that The Lancet study is based on “weak epidemiological data that relies on self-reported data”. These data are notoriously unreliable and shown to be correct only 0-20% of the time.

Robot-like analysis of low-carb diets

Feinman has taken his usual forensic scientific scalpel to what he calls its “questionable style of bloodless, robot-like computer analysis and acceptance of low hazard ratios” involved.

Hazard ratio (HR) is a term researchers commonly use in medical literature when describing survival data. They don’t only use survival data to describe the number of people who survive or die over a period. Increasingly, they also use these data to describe how many people can reach “a certain point in time without experiencing a hazard or event other than death (for example, suffering a heart attack) “. Or conversely, to determine the number that do experience an event.

The HR in The Lancet study was typically below 1.5 that Feinman says makes it “uniquely absurd”. (Some scientists say that a hazard ratio of above 2 is the barest minimum required for any robust conclusions. Others set the bar far higher.)

Click here to read: Teicholz: How low-fat diets can kill you


The Lancet researchers have characterized subjects’ nutritional status on the basis of two food questionnaires separated by six years and an HR of 1.12. This further “violates standards of research integrity”, Feinman says.

He identifies another fatal flaw. The researchers have claimed that low-carb diets are dangerous without actually studying low-carb diets.

“You cannot say that low-carb diets are bad if you have not measured low-carbohydrate diets,” Feinman says. Low-carb diets are already well-established and “do not allow setting up your own straw-man”.

In addition, if you criticize diet or any theory, you are obligated to present the opposing point of view and explain what is wrong with them, he says.

Irresponsible global headlines

But the most likely “landmark” that makes this study fatally flawed is the claim “that low-carbohydrate diets are life-threatening”, Feinman says. The researchers have made it on the basis of “dreadful analysis”.

Dr Zoë Harcombe

British obesity and public health researcher Dr Zoë Harcombe says that the study has generated “irresponsible global headlines”. In an article on her website, Harcombe highlights the study flaws behind the headlines.

Epidemics of obesity and type 2 diabetes show no sign of abating, she says. Therefore, it is “unhelpful, to say the least,” to encourage consumption of at least half of one’s diet in the form of the one non-essential macronutrient (carbs).

One of the flaws she identifies is that the researchers have chosen groups for carb intake and consumption subjectively. “Not even the carb ranges are even,” she says.

Most groups covered a 10% range (eg 40-50%). However, the range chosen for the “optimal” group (50-55%) was just 5% wide, Harcombe says. This placed as many as 6,097 people in one group and as few as 315 in another.

Massaging data

Thus, the subjective group divisions introduced what Harcombe calls “the small comparator group issue.” She explained the dynamic in the recent whole grains review. Harcombe repeats and builds on the explanation because she says it’s crucial to understanding The Lancet paper:

She gives as an example: If 20 children go skiing – 2 of them with autism – and 2 children die in an avalanche – 1 with autism and 1 without – the death rate for the non-autistic children is 1 in 18 (5.5%).  And the death rate for the autistic children is 1 in 2 (50%).

This demonstrates “how bad (or good) you can make things look with a small comparator group”, Harcombe says.

Click here to read: Naudé review: mistake or mischief against Noakes?


She also explains how the researchers managed to move from subjective grouping to the sensational life-expectancy headlines. They applied a statistical technique (called Kaplan-Meier estimates) to the ARIC data. “This is entirely a statistical exercise,” Harcombe writes. After all,  we don’t know when people will die. “We just know how many have died so far.”

This exercise resulted in the researchers’ estimates of longevity improvements:  They write that they “estimated that a 50-year-old participant with intake of less than 30% of energy from carbohydrate would have a projected life expectancy of 29·1 years. That’s compared with 33·1 years for a participant who consumed 50–55% of energy from carbohydrate.

Similarly, they estimated that a 50-year-old participant with high-carbohydrate intake (>65% of energy from carbohydrate) would have a projected life expectancy of 32.0 years. That’s compared with 33.1 years for a participant who consumed 50–55% of energy from carbohydrate.

Making mischief

 In this way, Harcombe says that the researchers used the small comparator group extremes (<30% and >65%) to “make the reference group look better”. It’s an example of “the mischief you can make when you subjectively make up groups”.

She returns to the example of the skiing children: If we were to use the data we have so far (50% of autistic children died and 5.5% of non-autistic children died) and to extrapolate this out to predict survival, life expectancy for the autistic children “would look catastrophic”.

This, Harcombe says, is exactly what has happened with the small groups – <30% carb and >65% carb – in The Lancet study.

In other words, she says that the researchers have manipulated the data. And as any good scientist will tell you, that’s just not on.

It would help, Teicholz suggests, if the media would “stop reporting on these kinds of (sensationalist) findings”. Unfortunately, that’s about as likely as most health journalists no longer drawing breath.