The Science of Human Potential

FGF-21, protein, carbohydrate, and mice

Once again, the significance of a mouse study has been distorted by its authors and exaggerated by the media.  It grabbed our attention because it was the lead front page story in our national newspaper the New Zealand Herald, no less. Also, because we’ve been big advocates for lower carb intakes, especially for the insulin resistant amongst us, it flies in the face on what we are on about.

So let’s have a look at what’s going on.


Common dietary advice has been turned on its head by new research finding that a low protein, high carbohydrate diet stimulates a hormone dubbed the “fountain of youth”.

The Sydney University group led by Dr Samantha Solon-Biet says a high-protein diet is good for reproductive health in younger adults, but recommends switching to a low-protein diet rich in vegetables and other natural carbohydrates from age 50 or 60 to live longer.

“A diet that optimises later-life health has a protein-to-carbohydrate ratio of 1:10,” Solon-Biet said.

The hormone this mouse study looks at is a protein called FGF-21. There is absolutely no evidence that this is a “fountain of youth” and really the myth of Ponce de León should be a warning to all these researchers – he never found the Fountain of Youth because it didn’t exist. Where this mouse research is interesting is because FGF-21, as well as having potentially useful effects in terms of insulin sensitivity, seems to be specifically important with regard to our appetite for sweet carbohydrates – when FGF-21 levels are high, carb cravings stop. So FGF-21 goes up when we eat carbs, so we know when to stop – but there’s such a thing as FGF-21 resistance, if levels are too high for too long we can become insensitive to it. FGF-21 levels also go up in other situations where we don’t need carbs – in prolonged fasting, or on a ketogenic diet. When you lose your sweet tooth after a few weeks on LCHF, that’s probably (in part) due to either a decrease in FGF-21 resistance or a rise in its level.

“FGF21 is increased in various conditions such as overfeeding, obesity, insulin resistance, starvation, protein/amino acid deprivation, low-protein ketogenic diets, and high carbohydrate feeding, yet the metabolic effects are very different in each nutritional context.”

But it’s very important to realise that thousands of proteins are involved in complex processes like the regulation of appetite. Maybe FGF-21 is to carbohydrate what leptin is to fat, but it’s very much early days still.

Auckland University Professor Wayne Cutfield said the Sydney study was based on mice, not humans, and looked at only one hormone called FGF21 out of a large number of hormones and other factors which interacted with each other.

“FGF21 is one of many factors involved in regulating our lifespan. To say that FGF21 is ‘the one’ is just not true,” he said.

Herald health and fitness commentator Lee-Anne Wann said people should not rush out and eat more carbohydrates to increase their FGF21.

“It’s great that they are looking at it. We’ve looked at other things like leptin which help manage our appetite, we are always finding new things,” she said. “But it’s too soon to be saying, ‘Oh my god, we should be increasing carbohydrates.”

In this instance the Herald reporter Simon Collins has done a good job (if we accept that these claims needed to be covered at all; but it will be big news in Australia for political reasons). The NZ experts he interviewed really don’t buy into the hype.

How was the study designed? In a somewhat confusing way. Mice were fed 25 different diets, and the paper isn’t clear what these were, except that they had various permutations of the 3 macronutrients. However, if they were the usual mouse diets, the carbohydrate would be mostly maltodextrin, the protein casein (just one of the proteins in milk, but the one which has a bit of a dodgy reputation when considered on its own). The fat could be anything – lard, cocoa butter, canola. Fortunately not much is being made of fat here.

But there’s a red flag – the mouse diet was not “rich in vegetables and other natural carbohydrates”. In fact, the odds are very high that it actually supplied none.

How were the results presented? Again in a somewhat confusing way. These scientists like using a novel kind of data visualisation, but fortunately they did provide this clearer summary.

Here’s your Fountain of Youth.


Eating a protein restricted, high carbohydrate diet maximises FGF21, but this results in an increased appetite for protein, for which FGF21 may be responsible.
If FGF21 was the Fountain of Youth, then, to exploit this, you would need to eat in such a way that you would always be craving protein, and then, if you did eat enough protein, you’d lose your FGF21 advantage.

This does not sound like much fun, even for a mouse. Protein foods are flavoursome, come in considerable variety, and are excellent sources of vitamins, minerals, essential fats, and energy. The LCHF diet isn’t, and the Paleo diet needn’t be, high protein at all, but for humans there is a real advantage to keeping muscle on our bodies (mice, for example, don’t need to open jars or lift the furniture often). This helps us to stay active, makes us more useful in our daily lives, and helps with physical and mental health. Losing muscle because of some fad diet idea (sic) that involves restricting protein for longevity (rather than for some measurable shorter term benefit, which you can actually check) is perhaps not the first thing that scientists should promote in the media. Surely every mouse experiment doesn’t have to generate a headline – there are enough human experiments to be going on with.

But I guess “high carbs key to long life” in a headline is just irresistible.

The discussion part of the paper is indeed interesting, and at least some of these 20 researchers really do know their stuff, but some of them seem to be extrapolating from it in the media as evidence that one type of traditional human diet out of thousands is healthier than all the others. The facts are, that

  • all traditional, pre-industrial diets are more-or-less healthy except when resources are poor.
  • populations eating diets high in meat and cheese (Sardinia) can be as long-lived as populations eating yams with low protein (Okinawa), and there are also shorter-lived populations eating both types of diet.
  • if we are no longer eating these traditional diets, we are more likely to have have metabolic disease or other health conditions that are rare in traditional populations, and which usually respond to specific and perhaps novel dietary changes – even peasants emigrating from the Mediterranean 100 years ago were advised to stop eating carbohydrate foods if they became diabetic in their new homes. Such changes can reasonably be expected to improve longevity for the individual more often than not.

The article also mentioned the Mediterranean diet. This is not actually low in protein, nor high in carbohydrate, as far as we know. A new EPIC study shows that a high Mediterranean Diet Score (they tested 4 different Med Diet scoring systems) is associated with lower mortality in an English population (i.e. compared to the normal English diet, one shudders to think what that is today). One interesting finding is the cholesterol counts of those with low and high Med Diet Scores. Mean total cholesterol in both groups is 6.2 mmol/L, LDL is 4 and 3.9 respectively, HDL is 1.4 and 1.5, TGs are 1.9 and 1.7.
So even the high Mediterranean Diet Score consumers in the UK tend to have high cholesterol, and the diet doesn’t make a lot of difference to that, yet it reduces cardiovascular and total mortality. Interesting.

A few years ago researchers showed that cell phone radiation cures Alzheimer’s Disease in mice. I don’t remember those researchers going all over the media telling us to strap cell phones to our heads from age 50 or 60 to think better. What is it about diet researchers?


[1] Solon-Biet et al., Defining the Nutritional and Metabolic Context of FGF21 Using the Geometric Framework, Cell Metabolism (2016),

[2] Tong TYN, Wareham NJ, Khaw K-T, Imamura F, Forouhi NG.
Prospective association of the Mediterranean diet with cardiovascular
disease incidence and mortality and its population impact in a non-Mediterranean
population: the EPIC-Norfolk study.
Tong et al. BMC Medicine (2016) 14:135
DOI 10.1186/s12916-016-0677-4



Can a high-fat diet protect against statin side-effects?

In the news recently is a review of statin benefits and side-effects in the Lancet which, using a controversial modelling method to predict population effects from the variable results of clinical trials, recommends that statins be prescribed more widely to healthy people to lower their risk of future heart attacks.[1]

The claim is that side-effects are rare, but they seem to be more common, and more serious, in real life than Professor Rory Collins, lead author of the review, admitted when interviewed by the BBC. In our opinion, it’s not ethical to try to trivialise the side effects of any drug that can kill or cripple the people taking it, as Professor Collins did, whether or not the benefits outweigh the risks at a population level.

It’s also worth taking into account how these statin trials are typically designed and what that means for side effects, exclusion and interpretation. We’ve written about that before here.

A commenter on the What The Fat blog sent us this link to an Official Information Act request about the number of statin-related deaths and injuries in New Zealand.

Over the period 01 January 2001 through to 31 December 2014, this document details 1709 reports describing 3826 reaction terms – one report may have more than one reaction described. 21 cases resulted in an outcome of death however in 3 cases death was not related to the Statin medicine and 2 cases were unclassifiable.

So we have 1709 reports of statins causing injury, and 21 deaths of which 15 were considered to be caused by statins. No doubt there is considerable underreporting here, as many people prescribed a drug which produces adverse effects will stop taking it without telling the doctor, and many doctors will stop prescribing such a drug without reporting the incident, so these numbers will tend to represent serious cases that weren’t easily resolved.

Why do some people experience extreme toxic reactions from statins while others tolerate them? A simple explanation for some types of harm might be, that statins reduce the synthesis of cholesterol, which lowers LDL in the blood, but everyone with a raised LDL cholesterol who is prescribed statins may not have an excessive rate of cholesterol synthesis to begin with, as cholesterol synthesis is regulated by insulin.[2] Someone with hyperinsulinaemia due to carbohydrate intolerance will tend to have an increased cholesterol synthesis, but this will often be accompanied by low or normal LDL levels. This may be why, in modern guidelines, LDL is no longer used as the sole guide to statin prescribing (the TG/HDL ratio is a better guide to insulin status).

Potentially, statins could cause enough of a cholesterol deficiency in cells to cause harm. This is how the NASA doctor Duane Graveline, who died recently, explained his own adverse reaction to statins, which caused him to suffer from amnesia. Professor Collins denies that amnesia is caused by statins; however all cases of amnesia in the New Zealand report (39) relate to the two statins, atorvastatin (Lipitor) and simvastatin (Zocor) which are fat-soluble and cross the blood brain barrier, just as Duane Graveline predicted.[3] Of course it is likely that amnesia in elderly patients prescribed statins is often missed as a drug side-effect, with resulting under-reporting.

A more complex mechanism for harm is that statins can be metabolised to lactones in some people, and these statin lactones are three times more toxic than the statins themselves, especially to muscle cells. Statin lactones inhibit mitochondrial respiratory complex III, reducing its activity by 84% in this experimental paper, and smaller but significant reductions in CIII activity were found in muscle biopsies from patients suffering from statin myopathy.[4]

In conclusion, we demonstrate that the Qo site of respiratory CIII is inhibited by several statin lactones and provide evidence for an association between this off-target effect and statin-induced myopathies. Consequently, polymorphisms of UGTs, the enzymes converting statin acids into lactones, and CIII could be predisposing factors in statin-induced myopathies. We showed that both G3PDH and b-oxidation stimulation can prevent statin-induced respiratory inhibition, providing a rationale for therapeutic intervention.


CIII shown in context – the beta-oxidation fed electron transfer flavoprotein complex (ETF or CETF) that saves the day is, as usual, not shown here.

CIII is part of the mitochondrial electron transfer complex or ETC (also known as the respiratory complex) which is part of the machinery cells use to generate ready-use energy units (ATP) from food. Its inhibition has the effect of reducing the muscle cell’s ability to generate ATP, with apoptosis (self-destruction) of cells as an outcome. Fortunately the beta-oxidation step of fatty acid oxidation contributes to ATP through a separate mitochondrial transporter (not usually shown in diagrams, because “glucose is the most important energy source”), and this input is able to restore CIII activity – “Beta-oxidation also contributed to convergent electron flow into CIII (i.e. it stepped around the statin blockade) and reversed the effects, restoring CIII activity to 89% of normal”.

In this cell-culture experiment beta-oxidation was stimulated by adding palmitoyl-l-carnitine to the mixture; this is a molecule of saturated fat attached to a molecule of l-carnitine, which is the molecule that carries fatty acids into the mitochondria for beta-oxidation. Co-enzyme Q10 is the molecule that carries the electrons between complexes. It has been proposed that Co-enzyme Q10 and l-carnitine be used together to treat statin myopathy.[5] However, increased beta-oxidation is also something that happens naturally in people eating the LCHF diet – it’s called fat-burning. (on the other hand G3PDH, interestingly, is activated by fructose, but also by glycerol, part of the fat molecule).

A further way in which statins can cause damage, this time to brain and nerve cells, is by inhibiting the synthesis of vitamin K2. Many plant foods contain vitamin K1, which the body converts to K2 (mostly in the liver, but also in the brain).
When this occurs in the brain the MK4 form of K2 produced is an essential co-enzyme for the synthesis of special sulfur-containing lipids called sulfatides. Low CNS sulfatide levels are associated with cognitive decline and seen in the early stages of Alzheimer’s disease.[6]

The MK4 form of vitamin K2 is produced with the same enzyme (HMG-CoA reductase) that is targeted by statins.

From all these facts and hypotheses, we can arrive at a number of factors likely, at least in theory, to be protective to people using statins.

One is a fat-burning metabolism. Of course this is associated with low insulin levels and therefore unlikely to cause excessive cholesterol synthesis in the first place.
Another is l-carnitine. This is made in the body, but we get extra from animal foods, especially red meat. Another is co-enzyme Q10, we make this in the body but by using the same HMG-CoA reductase enzyme that makes cholesterol and vitamin K2. Co-Q10 is found in animal foods and also in vegetable oils.

Vitamin K2 is found in animal foods such as liver and eggs, in cheeses, and in other fermented foods such as natto and sauerkraut.
And cholesterol itself, of course, is only found in significant amounts in fatty animal foods, especially eggs, offal, and shellfish.

The irony is that the diet most likely to protect against statin side-effects (if that is in fact possible), is a high-fat diet with plenty of rich and tasty animal foods. Exactly the sort of diet you’ll be told to avoid by most of the doctors prescribing them.

We do have quite a few studies of carbohydrate-restricted diets in people taking statins showing that the combination is safe and results in added improvement.[7]

In conclusion, these findings demonstrate that individuals undergoing statin therapy experience additional improvements in metabolic and vascular health from a 6 weeks CRD as evidenced by increased insulin sensitivity and resistance vessel endothelial function, and decreased blood pressure, triglycerides, and adhesion molecules.

And our favourite, the “high saturated fat and no-starch” diet.[8]

An HSF-SA diet was prescribed for all patients; they were instructed to attempt to consume one half of all calories as saturated fat, primarily as red meat and cheese. Eggs and other low-fat forms of protein were allowed, regardless of cholesterol content. Fresh fruit and nonstarchy vegetables were prescribed in restricted amounts at each meal.
Starch was forbidden.
In patients with atherosclerotic cardiovascular disease, an HSF-SA diet results in weight loss after 6 weeks without adverse effects on serum lipid levels verified by nuclear magnetic resonance, and further weight loss with a lipid-neutral effect may persist for up to 52 weeks. All patients were obese (mean +/- SD body mass index [BMI], 39.0+/-7.3 kg/m2) and had been treated with statins before entry in the trial.

This diet contains every element that (in theory at least) should protect against statin side-effects. It’s high enough in fat and low enough in carbohydrate to stimulate beta-oxidation (and even supplies some fructose to activate G3PDH), it supplies vitamin K2, l-carnitine, Co-enzyme Q10 and cholesterol.

Take home points

  • Statins inhibit the synthesis of cholesterol, excessive production of which is an effect of high insulin levels. Therefore, if statins do reduce heart attack risk, diets which lower insulin secretion by restricting carbohydrate should do so too. However, statins are only specific for one effect of excess insulin, whereas carbohydrate restriction reduces its effect on multiple pathways.
  • Statins can damage muscles, with life-altering consequences, by inhibiting the mitochondrial CIII complex involved in the production of energy (ATP). Diets which activate fat-burning in muscles by restricting carbohydrate can restore energy production (in theory, based on experimental evidence).
  • Statins can cause memory loss and neuropathy, probably by depleting vitamin K2 and cholesterol in the brain and nerves. High-fat, animal based diets are good sources of these nutrients, and also supply extra l-carnitine and Co-enzyme Q10 that may help to protect muscles.

As always, we are not cardiologists and are not qualified to prescribe statins or advise against using them. Those are decisions for your doctor and you (yes you!) to make when you weight up the likely benefits and harms and what weight to give to all of these combined.


[1] Interpretation of the evidence for the efficacy and safety of statin therapy.
Collins R, Reith C, Emberson J et al. Published Online September 8, 2016 S0140-6736(16)31357-5

Click to access PIIS0140-6736(16)31357-5.pdf

[2] Insulin and glucagon modulate hepatic 3-hydroxy-3-methylglutaryl-coenzyme A reductase activity by affecting immunoreactive protein levels.
Ness GC, Zhao Z, Wiggins L.
J Biol Chem. 1994 Nov 18;269(46):29168-72.

[3] Adverse Effects of Statin Drugs: a Physician Patient’s Perspective.
Graveline D.
Journal of American Physicians and Surgeons Volume 20 Number 1 Spring 2015

[4] Statin-Induced Myopathy Is Associated with Mitochondrial Complex III Inhibition.
Schirris TJJ, Renkema GH, Ritschel T, et al.
Cell Metabolism 22, 399–407, September 1, 2015.

[5] CoQ10 and L-carnitine for statin myalgia?
DiNicolantonio JJ.
Journal Expert Review of Cardiovascular Therapy. 2012: 10(10); 1329-1333


[6] Age- and brain region-specific effects of dietary vitamin K on myelin sulfatides.
Crivello NA, Casseus SL, Peterson JW, et al.
J Nutr Biochem. 2010 November; 21(11): 1083–1088.


[7] Dietary carbohydrate restriction improves insulin sensitivity, blood pressure, microvascular function, and cellular adhesion markers in individuals taking statins.
Ballard KD, Quann EE, Kupchak BR, et al.
Nutr Res. 2013 Nov;33(11):905-12. doi: 10.1016/j.nutres.2013.07.022. Epub 2013 Sep 18.

[8] Effect of a high saturated fat and no-starch diet on serum lipid subfractions in patients with documented atherosclerotic cardiovascular disease.
Hays JH1, DiSabatino A, Gorman RT et al.
Mayo Clin Proc. 2003 Nov;78(11):1331-6.




Response to Freedhoff and Hall – the differences between diets do matter.

In the latest Lancet, Yoni Freedhoff and Kevin Hall have an opinion piece about diet trials, in which they argue that it’s not helpful to know that an ad lib low-carb diet results in more weight loss than a calorie-restricted low-fat “dietary guidelines” type diet, because the weight loss differences are not clinically significant, a claim which we think is both false, and overlooking other benefits of carbohydrate restriction.[1]

They then go on to ask that more effort go into identifying ways to ensure that people can adhere to diets long term, a reasonable request and something we’re very much interested in ourselves. However, they spoil this a bit by talking about “hype” (short for hyperbole) and “society’s endless parade of fad diets”. If you want to improve adherence to diets and you think that there is little to choose between them, why use the word “fad”, which is normally used to devalue people’s dietary choices?
Further, is there even an endless parade of diets these days? Aren’t there a just few variations on and recombinations of the same timeless themes, such as vegan or vegetarianism, calorie restriction, carbohydrate restriction, ancestral, and so on? The only really original idea is fasting, yet this ancient therapy only seems original because we’ve just been led through such an intensive era of round-the-clock eating.

Freedhoff and Hall concentrate their argument on one trial, the DIRECT study (sometimes better known as Shai et al.) which was a 24-month workplace intervention in Israel, with ad lib low-carb, low-fat calorie restricted, and Mediterranean diet calorie restricted arms. “The low-carbohydrate, non–restricted-calorie diet aimed to provide 20 g of carbohydrates per day for the 2-month induction phase and immediately after religious holidays, with a gradual increase to a maximum of 120 g per day to maintain the weight loss…the [low carb] participants were counseled to choose vegetarian sources of fat and protein and to avoid trans fat. The diet was based on the Atkins diet.” The Mediterranean diet (at 40% fat, mainly from olive oil and nuts) was based on a 2001 book by Walter Willett and PJ Skerrett, and the low-fat diet was based on American Heart Association dietary guidelines. All dieters had access to the same food in the workplace cafeteria, but the food suitable for each different diet was colour coded.
In this study, people in the low carb group (not very low carb after induction, it varied between 87g/day at 6 months and 120g/day reported CHO intake over the first 12 months) lost on average 1.8 Kg more weight than the low fat dieters overall.[2]
Well as Freedhoff and Hall say, that’s not a lot of weight in the grand scheme of things. So does it support their claim that we should stop caring about the results of these studies?
Not so fast. The average weight loss includes all the people who drop out of the study; this is “intention to treat” (ITT) analysis, designed to keep the randomisation of baseline characteristics stable.
But what you might want to know if you were choosing a weight-loss diet, is, what will happen to me if I follow the diet? There were more drop-outs (22% vs 10%) in this study in the low carb arm, who found it hard to resist the biscuits and cakes in the cafeteria (interestingly, this caused their intake of saturated fat to increase over the study, even as their total fat intake went down), and self-reported complete adherence to low-carb was 57% at month 24.[3] The superiority of the low carb diet in DIRECT includes the effect of including this higher drop-out rate, and those extra cakes and biscuits. The per-protocol analysis only gives us a stratified comparison of completers vs non-completers (i.e. minus drop-outs, but including those with weak adherence to diets) at 24 months, but we do know from other studies that when non-completers are excluded, the long-term difference between diets at 12 months becomes larger.[4]


So compliance is important, sticking to the diet is critical of course, but what diet you stick to matters more than Freedhoff and Hall are saying. In the DIRECT study, people who completed 24 months of the low-carb diet lost a mean 5.5 +/- 7.0 Kg, and those who completed the low fat diet lost 3.3 +/- 4.1 Kg. The biggest loser in the DIRECT study lost 35% of their body weight, but all we know about the most successful dieters is, that they weighed more at baseline, lost weight more rapidly in the early stages, and ate a bit less protein and cholesterol at baseline (but overall the protein intakes in this population were, and remained, quite high). Rapid weight loss early in a diet is usually associated with success, and of course it’s a feature of the ketogenic diet, or the induction phase of the Atkins diet here.

But wait, there’s more. Weight loss isn’t the only effect of diet, and overweight people often suffer from increased cardiometabolic risk owing to insulin resistance and the metabolic syndrome.

Freedhoff and Hall for some reason don’t mention this, but it’s the evidence we have about the “long term safety” of any diet. In all parameters the ad lib low carb diet does better than the calorie-restricted low-fat diet, even at 24 months, and even including the drop-outs.

“Among the participants with diabetes, the proportion of glycated hemoglobin at 24 months decreased by 0.4±1.3% in the low-fat group, 0.5±1.1% in the Mediterranean-diet group, and 0.9±0.8% in the low-carbohydrate group. The changes were significant (P<0.05) only in the low-carbohydrate group (P=0.45 for the comparison among groups).”[2] In fact, as far as we know, carbohydrate-restricted diets are the only diets that can produce some of these benefits without weight loss.[5]

There’s a curious extra point in the chart below – LDL rose slightly at 6 months in the low-carb arm, when adherence was good and polyunsaturated fat intake was high, and dropped at 24 months when polyunsaturated fat intake decreased and carbohydrate, but also saturated fat, intake increased.

Shai markers

But wait, there’s more. Something else that Hall and Freedhoff didn’t mention is the very long-term effects of this diet trial, because there was a four-year follow-up study. And the results here are very interesting, because there is less of a rebound effect for the ad lib low carb diet and the Mediterranean diet than for the low-fat, calorie restricted AHA diet.[6]

Shai figure 1
In the low-fat diet group (which had the fewest drop-outs) most of the improved metabolic parameters, including weight, are back to baseline levels. In the low-carb group, weight and the LDL/HDL cholesterol ratio are still improved. This is four years after the end of a 2-year study – six years in all. Quite a different result from Hall’s Biggest Loser study, where severe rebound weight gain from CICO “eat less move more” energy restriction was the order of the day.[7]

After the completion of intervention, the participants were invited once a year to the clinic for a regular check-up and were encouraged to pursue a healthy diet. Although diet-group color coding and nutrition labeling in the workplace cafeteria were stopped at the end of the intervention, the cafeteria continued to serve suitable meals according to the guidelines of the 3 diets, suggesting that the workers could still consume their specific dishes, which continued to be regularly served, as they were during the trial. We did not continue with the dietary sessions or any other activity encouraging adherence. We used one question: Are you still dieting? The question had three possible answers to choose from: 1. “Yes, with my original diet” 2. “Yes, but I switched to another diet” 3. “No, I am not dieting”. No differences were observed in response to this question between the 3 assigned diet groups (p=0.36).


Perhaps there are a few people in this group who were so happy with their results that they stuck with the low-carb diet for 6 years, and their results are carrying the rest – or perhaps the 2 years of low-carb diet (or the 2 months of ketogenic dieting) had lasting benefits. These questions weren’t really answered by the questionnaire quoted above, which kind of refutes Freedhoff and Hall’s suggestion that we have nothing more to learn from diet comparisons. Good post hoc analysis of the data from weight loss trials can seek to develop further hypotheses about what baseline markers, characteristics, and responses predict success, and that should inform the design of further trials and interventions. It’s also possible to tweak the diets to improve them for both effect and ease of compliance – (e.g. what if the AHA diet had been lower-GI, the Med diet lower carb, and the low-carb diet lower carb, with more Mediterranean and real food elements? Something like this probably went on during the 4-year follow-up among the people still interested in the diets).
But in the meantime, we needn’t let ourselves get confused about how to proceed.

As Prof Richard Feinman says, “remind me again why we have a medical science literature?” Comparative trials of diets and drugs are designed and published so that we know what is the most effective option between two or more choices for any given diagnosis. The intention is that the best treatment, determined by experiment, will be the one to be offered first. The patient may not like it, or it may not work, in which case it will be time to try something else, but the evidence is there to inform the discussion.

Instead we are stuck in this Catch 22 where the evidence about the best treatment for overweight and diabetes, collected over decades at great expense, is ignored (or worse) because its results contradict cherished beliefs about (in this case) the pre-eminence and equivalence of the calorie, or (at other times) the health effects of saturated fat.

Fortunately some people are brave enough to follow the existing evidence while applying themselves to solving the question of adherence. David Unwin and colleagues in the UK have worked on the psychological aspects of motivating and supporting people in low carb diets for type 2 diabetes and NAFLD with great success,[8] and recently a multicenter LCHF approach in Canada has also reported good adherence and impressive results.[9]


A thought experiment

The DIRECT study had no control group, i.e. no group of people from the same population eating a normal (whatever they are normally eating)  diet ad lib.

Imagine there was a fourth arm randomised to an ad lib version of one of the 3 diets.
Without calorie restriction, it seems less likely that the AHA-approved, last-year’s dietary guidelines diet would have made any difference from baseline. It’s possible that nothing would have improved and plausible that things would have continued to get worse overall.
Without calorie restriction, it’s likely that the Mediterranean diet (the modern, updated dietary guidelines diet) would still have been better than the AHA-approved, last-year’s model dietary guidelines diet.There may well have been some smaller improvements, and things would be unlikely to get worse.

Now imagine an ad lib version of the LCHF diet (the controversial, alternative-dietary guidelines diet). The results would still be exactly the same, because the experimental diet was ad lib.

[1] Freedhoff Y, Hall KD. Weight loss diet studies: we need help not hype. The Lancet , Volume 388 , Issue 10047 , 849 – 851.

[2] Shai, I, Schwarzfuchs, D, Henkin, Y et al. Weight loss with a low-carbohydrate, Mediterranean, or low-fat diet. N Engl J Med. 2008; 359: 229–241

[3] Greenberg, I, Stampfer, MJ, Schwarzfuchs, D, and Shai, I. Adherence and success in long-term weight loss diets: the dietary intervention randomized controlled trial (DIRECT). J Am Coll Nutr. 2009; 28: 159–168

[4] Feinman RD. Intention-to-treat. What is the question? Nutr Metab (Lond). 2009 Jan 9;6:1. doi: 10.1186/1743-7075-6-1. Full text:

[5] Gannon MC, Nuttall FQ. Control of blood glucose in type 2 diabetes without weight loss by modification of diet composition.  Nutrition & Metabolism 2006 3:16 DOI: 10.1186/1743-7075-3-16. Full text:

[6] Schwarzfuchs D, Golan R, Shai I. Four-year follow-up after two-year dietary interventions. N Engl J Med. 2012 Oct 4;367(14):1373-4. doi: 10.1056/NEJMc1204792.
Full text:

[7] Fothergill, E, Guo, J, Howard, L et al. Persistent metabolic adaptation 6 years after “The Biggest Loser” competition. Obesity (Silver Spring). 2016; DOI: (published online May 2.)

[8] Unwin DJ, Cuthbertson DJ, Feinman R, Sprung VS (2015) A pilot study to explore the role of a low-carbohydrate intervention to improve GGT levels and HbA1c. Diabesity in Practice 4: 102–8. Full text:

[9] Mark S, Du Toit S, Noakes TD, Nordli K, Coetzee D, Makin M, Van der Spuy S, Frey J, Wortman J. A successful lifestyle intervention model replicated in diverse clinical settings. S Afr Med J. 2016 Jul 3;106(8):763-6. doi: 10.7196/SAMJ.2016.v106i8.10136.
full text:

Australia’s response to the diabetes epidemic – shooting the messenger.


On the Sunday current affairs program in New Zealand there was a report on the diabetes epidemic in South Auckland. This is our largest (and growing) health problem, and two of the players in this tragedy had messages that stood out. An elderly woman, overweight and now condemned to thrice-weekly dialysis, told us “I didn’t do anything wrong”.

How right she was. The Ministry of Health website still offers this “healthy eating” advice – “Fill up on breads, cereals, pasta and rice.”

Junk epidemiology and junk food

The epidemiologists from Harvard recently grabbed headlines with claims that polyunsaturated fats are the healthiest fats, that chicken is one of the healthiest animal proteins, and that plant proteins are healthier than animal proteins. In South Auckland, a staple food is Kentucky Fried Chicken. Chicken is a meat naturally high in polyunsaturated fat, KFC is fried in “healthy” polyunsaturated vegetable oils, and therefore a good source of these, and it even comes with a bean salad – plant protein. This junk food meal, eaten wherever diabetes is rampant (the franchise only came to New Zealand in the 1970’s), actually ticks most of the boxes thrown up by junk epidemiology.

With this sort of dangerous misinformation on official government websites and in the media, how can anyone know what is “right” or “wrong” when it comes to their risk of type 2 diabetes?

The look of success?

One chap in the Sunday program, Kim, had it figured out. He’d reversed his diabetes by, firstly, losing weight rapidly on a low calorie diet (like the Newcastle diet, but with real food instead of Optifast), by exercising regularly, and by eating a diet described as “lots of vegetables” – we saw a delicious looking stir-fry – “no bread, potatoes, rice, pasta” – he didn’t even need to mention sugar.

Incidentally, what is a “fast-acting” carbohydrate? Previously, it was assumed that fast-acting carbohydrates were sugars or juices. It is now known that this is not true, and if there were a fast-acting carbohydrate, it would probably be a starch. “Fast-acting carbohydrate” is a term we need to eliminate from our diabetes vocabulary. – Marion J. Franz, MS, RD, LD, CDE

“So what’s left?” asked the interviewer. “Eggs, meat?” “You’d be surprised how much there is left that’s good to eat!”

Now, the direct end result of type 2 diabetes is a series of complications which include cardiovascular disease, retinopathy, neuropathy, kidney disease and gangrene (caused when vascular damage cuts of blood supply to the extremities, usually the feet, compounded by neuropathy preventing pain warning of injury, and infections fed by high sugar levels and suppressed immune function). Gangrene often requires the amputation of the affected parts, and the surgeon who has to perform this procedure is an orthopedic surgeon.

The case of Dr Gary Fettke

Gary Fettle is a friend of mine. He’s also an orthopedic surgeon in Tasmania, Australia who performs dozens of these operations on patients with type 2 diabetes every year. “I used to do one amputation every 6 to 12 months and now I’m doing one a week”. Seeing, as anyone can who has eyes in their head, the link between diet and diabetic blood sugars and the risk of complications (hardly controversial), he had dared to make a study of nutrition and diabetes literature – something which, as a highly trained medical professional, he was well able to do – and advise his patients, and the public, about how to eat to beat, and avoid type 2 diabetes and/or its complications.

Now Dr Fettke has been banned from giving any diet advice, to patients or in any media, until further notice.

Why? What he is advising is plainly good sense as well as evidence-based. It’s the same message Kim gave on the Sunday program, except that Dr Gary Fettke is a highly trained medical professional with hundreds of hours of clinical experience.

Here’s a great response to the silencing of Dr Fettke, written by Tyler Cartwright for the Ketogains website, that puts the case better than we can.

Meanwhile the Australian authorities continue to allow Associate Prof Sof Andrikopoulos to give diet advice, despite his telling the Australian public to eat sugar with burgers – based on his experience with mice. (I guess that also makes the soft drinks at KFC part of the healthy menu now).

Gary’s not the first and won’t be the last

In 2005 the Swedish dietetics authorities tried to silence Dr Annika Dahlqvist. Their heavy-handed actions led to a court case in 2008 which Dr Dahlqvist won, publicising the benefits of LCHF all over Sweden, and as a result  a significant proportion of the Swedish population soon knew about the diet, and butter sales went up – leading to much hand-wringing around the world among people committed to outdated bad advice, but no adverse effects in the Swedish population – according to the Swedish government’s health data base, heart attacks are now at an all-time low.

20-85+ 39,418 38,846 37,150 34,780 34,140 32,814 32,149 29,823 28,783

Heart attacks in Sweden by year, 2006-2014.



In this recent Australian TV series, The Saving Australia Diet, Dr Fettke is seen advising the patient Tony on how to treat diabetes with the LCHF diet, with the help of chef Pete Evans. For no good reason that we can see, other than some virulent local strain of the Tall Poppy syndrome, the Australian establishment hates Pete Evans, and this has made some scientists who should know better indulge in bottom-of-the-barrel stunts like Ass Prof Sof Andrikopoulos’s “Paleo mouse” attacks on low carbohydrate diets. It is almost certainly his association with Pete Evans that has drawn the complaint that has led to Dr Fettke being silenced.

Of course, this kind of heavy-handed, bloody-minded action is only possible because Dr Fettke is a health professional, and therefore subject to the discipline of a regulatory body, even if it is being abused for unworthy personal ends and is clearly not in the public interest. Pete Evans, on the other hand, is a member of no such body, so he can’t be silenced, thank goodness.

This is why it’s important for everyone who speaks on nutrition to have a proper qualification – so they can be silenced when they embarrass the authorities, for example by being right about something the government and its appointed experts have been consistently wrong about. Especially in the middle of an epidemic, when damage control is the order of the day.

Well, here’s an idea – instead of “damage control” being about saving reputations, can’t we have damage control that will mean saving feet, eyes, and kidneys?

We don’t always say good things about Aussies (us New Zealanders, and vice versa), but they are our mates really and Gary Fettle is one of the good ones.  Shame on you Australia and the Australian Medical Authorities for allowing this to happen.

Intro to low carb and fasting seminar

MIL154729 HPC Facebook Post (1200 x 1200)We are doing a seminar Sept 8th, AUT Millennium Auckland

Presenters – me (Grant Schofield), dietician Dr Caryn Zinn, and Jimmy Moore all the way form the USA talking about his experiences with fasting.

It’s definitely an intro night, so well suited to those just getitng into or supporting others getting into this lifestyle

Limited space – book online here

  • Date: Thursday 8th September
  • Location: AUT Millennium, The Finish Line,
  • 17 Antares Place,Mairangi Bay, Auckland
  • Time: 6.30pm –8.30 pm
  • Tickets: $25, limited to the first 150 people
  • Buy your tickets online at
  • Sorry no door sales

PDF flyer here MIL154729 HPC Flyer 8.9.16 2



The “thrifty” gene found in Samoa and what it means

SamoaThis (very technical) paper has made a big splash on the internet and in the media, not just in New Zealand but all around the world.

A thrifty variant in CREBRF strongly influences body mass index in Samoans
Ryan L Minster, Nicola L Hawley, Chi-Ting Su, Guangyun Sun, Erin E Kershaw, Hong Cheng, Olive D Buhule, Jerome Lin, Muagututi‘a Sefuiva Reupena, Satupa‘itea Viali, John Tuitele, Take Naseri, Zsolt Urban, Ranjan Deka, Daniel E Weeks, & Stephen T McGarvey
Nat Genet. 2016 Jul 25. doi: 10.1038/ng.3620. [Epub ahead of print]

Samoans are a unique founder population with a high prevalence of obesity, making them well suited for identifying new genetic contributors to obesity. We conducted a genome-wide association study (GWAS) in 3,072 Samoans, discovered a variant, rs12513649, strongly associated with body mass index (BMI) (P = 5.3 × 10−14), and replicated the association in 2,102 additional Samoans (P = 1.2 × 10−9). Targeted sequencing identified a strongly associated missense variant, rs373863828 (p.Arg457Gln), in CREBRF (meta P = 1.4 × 10−20). Although this variant is extremely rare in other populations, it is common in Samoans (frequency of 0.259), with an effect size much larger than that of any other known common BMI risk variant (1.36–1.45 kg/m2per copy of the risk-associated allele). In comparison to wild-type CREBRF, the Arg457Gln variant when overexpressed selectively decreased energy use and increased fat storage in an adipocyte cell model. These data, in combination with evidence of positive selection of the allele encoding p.Arg457Gln, support a ‘thrifty’ variant hypothesis as a factor in human obesity.

Samoa gene

How the variant gene correlates with BMI in Samoa

This is a big deal because the thrifty gene hypothesis – that some people are prone to obesity because their ancestors evolved to store energy more effectively in times of plenty – had been dead in the water for lack of convincing evidence. Genes are complicated things and previous studies hadn’t turned up anything with this sort of effect size (obesity is common today, yet these “thrifty” genes are normally quite rare). The thinking is that the gene variant (technically a “missense” mutation where a particular amino acid in a protein has been replaced with a different amino acid that reduces its function) was selected for during long sea voyages, where starvation and hypothermia weeded out individuals who had been less effective at storing and/or holding onto fat.

These genes are not destiny

Stephen McGarvey, the senior author, was quick to point out in all his interviews that this genetic evidence doesn’t mean that such genetics destine Pasifika people to be obese independently of environment, that is, diet, lifestyle, and exercise. (Though the study only looked at Samoa, it is almost certain that the gene will be found to be similarly common in other Pasifika populations). It is unlikely that the va’a tele or waka that carried the Polynesian explorers round the Pacific basin carried any obese individuals even when they set out; the stored energy advantage that the gene variant gave in these times was probably a few Kg and barely noticeable. If we look online for historical photographs of Samoa, we find that Samoa has a rich political history – it was the meeting place of the German, British, and U.S. Empires in the late 19th century – which resulted in much photographic documentation. But even as recently as 1930, during the period of the Mau independence movement, there is no evidence of obesity in photos.
These are pupils of Vaipouli school, a Samoan secondary school, ca 1930.


And these (photo at top) are leaders of the Mau movement at the same time.


Regardless of genetics, indigenous populations rarely if ever experience obesity eating their traditional diets, and always experience a high rate of obesity and/or diabetes, as well as the other diseases of Western civilisation (tooth decay, heart disease, cancer, appendicitis), after transitioning to Western processed and refined foods – foods with a high Human Interference (HI) factor. Traditional Samoan staple foods are taro and coconut, breadfruit and banana, turtle, pork and fish; this is never a question of fat vs carbs, or of saturated fat vs unsaturated fat, nor even of fructose vs glucose (the Ti root prepared in much of polynesia is very high in fructose) – it’s a question of natural foods with their fibre, micronutrients, fats and protein intact, vs refined starch, sugar, and refined oils, including the soybean oil used to pack canned fish in the case of Polynesia.[1]

Obesity prevents diabetes? Come again?

A really interesting finding from this study from our point of view is that the “obesity” gene variant is strongly protective against type 2 diabetes.

Higher BMI and adiposity are usually associated with greater insulin resistance (higher fasting insulin levels and homeostatic model assessment–insulin resistance (HOMA-IR)), an atherogenic lipid profile (especially higher serum triglyceride and lower HDL cholesterol levels), and lower adiponectin levels. We therefore expected the BMI-increasing A allele of rs373863828 to also be associated with these metabolic variables. However, even though the A allele was consistently associated with higher BMI and adiposity in both the discovery and replication cohorts, the expected associations with the above obesity-related comorbidities were not observed and, in some cases, were even in the opposite direction to that expected (Table 2 and Supplementary Table 2).
Notably, when considering all subjects, the risk of diabetes was actually lower (OR = 0.586 for the discovery cohort, P = 6.68 × 10−9) or trended lower (0.742 for the replication cohorts, P = 0.029) in carriers of the A allele. Likewise, even in non-diabetic subjects, the variant was associated with moderately but significantly lower fasting glucose levels in both the discovery and replication cohorts (1.65 mg/dl (P = 9.5 × 10−5) and 1.54 mg/dl (P = 8.8 × 10−4) lower for each copy of the A allele, respectively). These effects became even more significant after adjusting for BMI (2.25 mg/dl, P = 6.9 × 10−8 and 2.09 mg/dl, P = 7.6 × 10−6), suggesting an independent effect of the variant on glucose homeostasis and diabetes risk.

How can this be? Obesity is associated with, and plausibly causal in, diabetes, yet a gene variant that increases obesity halves the rate of diabetes?
This is in fact strongly supportive of the ectopic fat hypothesis of diabetes causation, and the personal fat threshold hypothesis of Professor Roy Taylor of Newcastle Diet fame.

“We hypothesize that each individual has a personal fat threshold (PFT) which, if exceeded, makes likely the development of T2DM. Subsequent weight loss to take the individual below their level of susceptibility should allow return to normal glucose control. Crucially, the hypothesized PFT is independent of BMI.”[2]

“A key reason for the greater increase in diabetes risk per unit increase in BMI in South Asians compared with Europeans may be due to a reduced capacity in South Asians to store fat in the primary superficial subcutaneous adipose tissue compartment, leading to earlier `overflow’ into secondary deep subcutaneous and visceral fat compartments, and potentially the liver.” [3]


(Professor Taylor’s research also pinpoints the pancreas as a target for this “overflow”).
Imagine the body as a house in which rainwater (energy stored as fat) is collected in a tank; if the water is not used, and the tank fills, the water will overflow into the walls and floors of the house; the house will be at risk of rotting and the electrics might short out. The larger the tank, the less risk of damage there will be in a downpour. But no matter how big the tank, the house can still be damaged if it fills to overflowing. This tank is the capacity to store energy (fat, and the fat produced from carbohydrate) subcutaneously (in the folds of fatty tissue under the skin).
So it is not a paradox that while weight gain increases the risk of type 2 diabetes, some of the people who are most prone to obesity for genetic reasons will have a lower risk of type 2 diabetes at a given BMI.

Take-home messages:

1) about 1 in 4 people in Samoa, and possibly similar numbers in other Pacific populations, have a variant gene that makes it easier to store fat and become obese. This strengthens the “thrifty gene” hypothesis of obesity, at least with regard to these populations.

2) This gene is also associated with a significantly reduced risk of diabetes, a fact which strengthens the “ectopic fat” and “personal fat threshold” hypotheses of type 2 diabetes.

3) Variant genes or their lack did not result in obesity or diabetes in Samoan populations living their traditional lifestyles and eating their traditional diets of unrefined, nutritious animal and vegetable foods. Western processed foods and refined high-energy ingredients have been a disaster for these peoples. This strengthens the “gene – environment mismatch” theory of modern diseases, and the Paleolithic diet hypothesis.


[1] Behavioral risk factors for obesity during health transition in Vanuatu, South Pacific.
Kelsey Needham Dancause, Miguel Vilar, Michelle Wilson, Laura E Soloway, Christa DeHuff, Chim Chan, Len Tarivonda, Ralph Regenvanu, Akira Kaneko, J Koji Lum, and Ralph M Garruto. Obesity (Silver Spring). 2013 Jan; 21(1): E98–E104. (full-text link)

[2] Normal weight individuals who develop Type 2 diabetes: the personal fat threshold. Roy Taylor, Rury R. Holman. Clinical Science Apr 01, 2015, 128 (7) 405-410; DOI: 10.1042/CS20140553

[3] Type 2 diabetes as a disease of ectopic fat?
Naveed Sattar and Jason MR Gill. BMC Medicine 2014:12(123) (full-text link)

Eating more saturated fats raises risk of early death; yeah, right.


Behold, a new diet-health observational paper appears! (What, this one, again?)

There are two long term observational studies of diet and health that get more press coverage than all the others combined. New papers on these studies, which say the same things with minor variations, are published every few months in high-impact journals. The editors of these journals don’t seem to mind that the papers they get are unoriginal and the repetitive choice of the same studies amounts to cherry picking, because the authors are the rock stars of the nutritional establishment, and their university, Harvard, has a cachet that guarantees publication, just as an education at Eton guarantees influence in the British political establishment no matter how much of an idiot one happens to be.

Was that too critcal? You be the judge.

The studies are the Nurses Health Study (NHS), which had run for 30 years and is all-women (being a male nurse must have been an exclusion criteria for this study) and the Health Professionals Follow-up Study (HPFS) which had run for 26 years (and, you guessed it, included no female health professionals) when the results were analysed yet again to create this new paper.

Association of Specific Dietary Fats With Total and Cause-Specific Mortality
Dong D.Wang, MD, MSc; Yanping Li, PhD; Stephanie E. Chiuve, ScD; Meir J. Stampfer, MD, DrPH; JoAnn E. Manson, MD, DrPH; Eric B. Rimm, ScD; Walter C. Willett, MD, DrPH; Frank B. Hu, MD, PhD.
JAMA Intern Med. Published online July 05, 2016. doi:10.1001/jamainternmed.2016.2417

The mortality rate in these two studies, given their long duration, was quite high – 33 304 deaths among 126233 people. After adjustment for known and suspected risk factors, dietary total fat compared with total carbohydrates was inversely associated with total mortality (hazard ratio [HR] comparing extreme quintiles, 0.84; 95% CI, 0.81-0.88; P < .001 for trend). The HRs of total mortality comparing extreme quintiles of specific dietary fats were 1.08 (95% CI, 1.03-1.14) for saturated fat, 0.81 (95% CI, 0.78-0.84) for polyunsaturated fatty acid (PUFA), 0.89 (95% CI, 0.84-0.94) for monounsaturated fatty acid (MUFA), and 1.13 (95% CI, 1.07-1.18) for trans-fat (P < .001 for trend for all).

Taken at face value, this tells us that people who ate a higher-fat, lower-carb diet overall were less likely to die, over three decades, than people who didn’t. The amount of saturated fat people ate didn’t have much, if anything, to do with this.

So how was this reported in the media?

People who eat more saturated fat have a higher risk of an early death, according to a large study that contradicts recent claims that “butter is back”.

The study from the Harvard TH Chan School of Public Health in the United States has been following 126,000 people for three decades to assess the impact of their diet on their health and lifespan. The researchers claim it is the most detailed and powerful examination to date of the effects of eating different types of fats.

Its findings run counter to those of a study published by the National Obesity Forum in May, which said people should eat more fat and fewer carbohydrates and rubbished the nutritional guidelines from Public Health England, which recommend that people should eat less butter and red meat.

According, that is, to The Guardian. Last time we read the National Obesity Forum document, it was recommending that people at risk of diabetes and obesity eat more fat, mainly from wholefoods, while using olive oil and butter for cooking, and restrict carbohydrates, especially refined carbohydrates.

This is a dietary pattern that, if eaten by the people in the NHS and HPFS, would have increased fat in place of carbs (HR 0.84), increased MUFA (HR 0.89), increased PUFA (0.81), and decreased trans fats (HR 1.13). In other words, following the National Obesity Forum recommendations should have significantly decreased mortality. In this context, the HR 1.08 for SFA (a tiny 8% increased risk) is relatively meaningless. We’ll have a look at how it was arrived at next (spoiler, it was arrived at poorly).
But first it’s worth talking about trans fats. Trans fats are formed when unsaturated vegetable oils are partially hardened by hydrogenation to create spreads and shortenings that can be used in place of animal fats like butter, lard, and tallow. These industrial trans fats (mainly elaidic acid, the trans isomer of oleic acid) are associated with an increased risk of CVD mortality at relatively small intakes and are probably best considered toxic in all but naturally occurring trace amounts.
Industrial trans fats only exist in the food supply because of the widespread fear of natural saturated fat and animal fats. The people who warn us against trans fats today include some of the same people who made us eat them in the first place.

How does the Guardian article get around this inconvenient truth? By treating trans fats and saturated fats as if they are part of the same problem.

The new paper, published in the prestigious US journal Jama Internal Medicine, says eating trans-fats and saturated fats, including those from red meat and butter, is linked to higher mortality rates compared with the same number of calories from carbohydrates.

More importantly, they say, they found that death rates dropped by between 11% and 19% among people who substituted saturated fats such as butter, lard and red meat for unsaturated fats such as olive oil, canola oil and soybean oil.

The first claim is false; the trans fats in red meat and dairy are widely believed to be beneficial and weren’t the trans fats measured in the studies. Let’s look at that latter claim – it might surprise you to know that no-one in this study substituted any food for any other food (or rather, if they did, this wasn’t looked at). This is all imagination; and when these substitutions were tested for real, they didn’t result in lower mortality. Also, there’s a lot of unsaturated fat in butter, meat and lard. The unsaturated fat that’s supposed to be good for people in this study. It’s also important to note that Americans eat very little butter (it was 4.8% of total fat in 1955, before the scare started); the contribution of butter and lard to saturated fat intake in these studies would have been small, and in a recent meta-analysis of butter alone, the first ever done, it has zero correlation with cardiovascular disease.

How were the data analysed?

But how were the results obtained? (If interested, you can find detailed, and somewhat scathing, critiques, by other experts, of the methods used here on the PubPeer website, which go into much more depth). Participants in the study were asked, every few years, to remember what food items they had eaten in the previous year. Really. “In each SFFQ, we asked how often, on average, the participant had consumed a specified portion size of each food during the preceding year.” Oh dear. The NHS participants have reported, by this method, eating only an average ~1,500 kcal a day for 32 years. For the purposes of our analysis, we’ll have to reluctantly take these data at face value, but seriously? (an ongoing problem in nutrition research really, especially these types of cohort studies)

In the raw NHS data, women in the lowest quintile for saturated fat intake (average age 48.2 years) had more than double the death rate of women in the highest quintile (average age 45.9 years) (5852 vs 2332). So saturated fat intake was spread unevenly by age; younger women ate more of it (and people with low cholesterol ate more); or so it seems from the raw data (though the paper says otherwise); but would an extra 2.3 years really make such a huge difference? When both studies were combined, and the deaths were adjusted for age, the HR for saturated fat was 1.71 (1.65-1.78). This is quite a correlation – but there were a lot of differences, besides age, between the upper and lower quintiles for saturated fat intake. More smokers and less exercise in the upper quintile for saturated fat, for instance. When the data were adjusted for “known and suspected risk factors” the correlation was bumped down to the minimal HR of 1.08. The difference between 1.08 and 1.71 is +0.63. That is, there are things here that are associated with a 63% increase in mortality, and the authors had to pare those away to arrive at something associated with an 8% increase, which is what they decided to warn us about.
I don’t know about you, but I’m more interested in this 63% risk, which dwarfs anything any of these fats might be doing to us.

After age, these adjustments were made; this was the 63%.

Adjusted for white race, marital status, body mass index, physical activity, smoking status, alcohol consumption, multivitamin use, vitamin E supplement use, current aspirin use, family history of myocardial infarction, family history of diabetes, family history of cancer, history of hypertension, history of hypercholesterolemia, intakes of total energy and dietary cholesterol, percentage of energy intake from dietary protein, and menopausal status and hormone use in women.


Smoking  (for example) wasn’t evenly distributed among quintiles, and passive smoking, which was among the many possible “known and suspected” risk factors not measured, probably has a similar distribution to smoking. So does sugar consumption; though measured in these studies and previously associated with both saturated fat consumption and CVD mortality, this also wasn’t adjusted for. What’s interesting here is that these authors think that vitamin E supplement use is worth adjusting for. Polyunsaturated vegetable oils are the main source of vitamin E in modern SAD-type diets.

Why do smoking, inactivity, and no doubt sugar consumption and other unhealthy behaviours associate with saturated fat intake? After all, there’s no natural explanation for this. Steak doesn’t make you crave cigarettes, coconut oil doesn’t make people stop exercising. But this is a study involving two groups of health professionals, followed during the years 1980-2012. These people were TOLD, over and over again, that foods that were sources of saturated fat were unhealthy and foods that were sources of polyunsaturated fat were healthy.  We ask whether it’s ever possible to do statistical adjustments to model the independent associations of single variables (e.g., saturated fat) by various statistical techniques, given the public health messaging over the decades which inextricably confounds these results?
The people who cared about being healthy ate less butter and red meat, and used polyunsaturated oils and spreads. They exercised more and probably did many other things we don’t measure well or even at all. The people who couldn’t be bothered made less of these changes, were more likely to keep smoking and care less about second-hand smoke, probably drank too much and for all we know were more likely to have unprotected sex and use drugs – and on the list can go. They included the risk-takers, the contrarian ignorers of advice, the fun-lovers, and the pessimists and fatalists. This is called “healthy user bias”, and it’s always a problem when analysing observational studies of behaviours that society has strong opinions about.

Can we show that healthy user bias or residual confounding or the considerable potential for error built into the SFFQ system accounts for the 1.08, the remaining 8% of risk not seen in better quality studies of less easily biased populations? Saturated fat is believed to increase heart disease risk (the largest part of CVD mortality) by raising LDL cholesterol, you know the story. This is not a mechanism that anyone thinks plays a role in other causes of death; yet the Harvard study also found a positive correlation between respiratory disease mortality and saturated fat (HR 1.56; 95% CI, 1.30-1.87) and an inverse correlation with PUFA. This is a condition we’d expect to relate directly to smoking and air quality, even more so than CVD. It is likely to be an indicator of residual confounding (the authors describe this association as “novel”, so there’s no support for it in any other studies). The study also found this same pattern of associations (SFA mildly bad, MUFA and PUFA good) with all of the diverse causes of death studied, including cancer and neurodegenerative diseases; this is impossible to explain in terms of the lipid hypothesis, but easy to explain if high saturated fat consumption in this Standard American Diet population was associated with lower micronutrient and antioxidant intake, greater risk taking (including exposure to infections), higher sugar and refined grain intake, and greater exposure to smoking and environmental pollutants.

This reminds us of the study that found red meat consumption was significantly associated with accidental death in men. So beef makes you clumsy, or drive too fast, or not wear a seatbelt. D’you think? It’s much more likely that this indicates some residual confounding around risk-taking behaviours that weren’t properly measured and accounted for, men being overall bigger risk-takers than women. Unfortunately, there’s no accidental death data in the Harvard paper to compare.

We also found an earlier 20-year follow up study from the NHS alone which found no correlation between saturated fat and heart disease incidence in this group – HR 0.97 (95% CI 0.73, 1.27).

Is canola oil consumption associated with cancer mortality?

If you look at supplement 1 for this new Harvard study, which gives the results for men (HPFS) and women (NHS) separately, they are sometimes quite different, and are less significant, than the correlations in the main paper. A really interesting result that’s in the supplementary data is that omega-3 alpha linolenic acid (ALA) consumption is significantly associated with cancer mortality.
The HR for cancer mortality between low and high quintiles of ALA intake (table 14, p62) is 1.12 (1.04, 1.20), which is greater than the HR between saturated fat and all-cause mortality. ALA is in healthy foods like leafy green veges, pulses, some nuts and seeds, and whole grains, in small amounts. It’s not likely that eating these increases the risk of cancer. But ALA is present in larger amounts in canola oil and soybean oil, and these are the main sources of ALA in the US diet.
Go figure, there could be many reasons for this, but it’s odd that this correlation, which runs against the grain of any likely healthy user bias or residual confounding, appears tucked away in the supplement to a paper which is being used to support the idea that we should replace butter and lard with canola spread and soybean oil.
The correlation is mentioned briefly (without figures) in the main paper, which tries to minimise it by saying that the correlation in the “most recent” fat consumption table (table 7) is non-significant. But surely the whole point of a 32-year study like this is to try to capture the effect of dietary factors over a significant part of the lifespan.

It’s predictable that Harvard are beginning to confirm the benefits of a high-fat diet (yes, even in the context of the confounders of this study), but annoying that they harp on the evils of the saturated fat found in naturally occurring and nutritious foods, and continue to promote industrially refined vegetable oils and spreads (which is how trans fats got in our diet in the first place), based on evidence collected by methods which don’t seem that reliable to begin with. In the USA they don’t require labelling of trans fats in oils and spreads and even in NZ this is still voluntary. Even going by Harvard’s own research into the NHS and HPFS cohorts, there’s a better way to get any health benefits of PUFA and vitamin E.

Overall ? Just eat real food with a mixture of fatty acids. Avoid trans-fats and advocate for more funding for intervention research, where cause and effect is more easily determined and confounding exists less because of randomisation.


The timing of carbs in meals makes a diference!



The spirit of science is not merely faith in the power of reason; it is also a belief that our problems may be simpler than they appear to be.

– Colin Wilson, The Strength to Dream

This recent pilot study is an example of good science for two reasons – it suggests that a minor change in behaviour can be a shortcut to important health benefits, and it raises more questions than it answers.

Alpana P. Shukla, Radu G. Iliescu, Catherine E. Thomas, and Louis J. Aronne.
Diabetes Care 2015;38:e98–e99 | DOI: 10.2337/dc15-0429

The authors looked at the effect of food order, using a typical Western meal (628 kcal:
55 g protein, 68 g carbohydrate, and 16 g fat), incorporating vegetables, protein, and carbohydrate, on postprandial glucose and insulin excursions in overweight/obese adults with type 2 diabetes. There were only 11 subjects (5 male, 6 female), which is why it’s a pilot study; the numbers were small, so any effect, to be statistically significant, actually had to be clinically significant too. But these people could be exceptions, the effect of metformin may be necessary for the response, it might vary when different foods are eaten, the effect might not be seen with larger meals, or in free-living populations, and the long-term effect on fasting glucose and insulin wasn’t investigated; so it’s a “proof of concept” type of experiment that tells us, loud and clear, that further investigations are certainly warranted.

The study was a cross-over design, meaning that both meal orders were tested in all subjects, a week apart. The meals were made up of ciabatta bread and orange juice (the carbohydrate part) and a chicken salad with low fat vinaigrette, plus broccoli and butter (the vegetable and protein part) and these two parts were eaten, separated by 15 minutes, in 2 different orders. Insulin and glucose were measured just before the meal and at 30, 60, and 120 minutes after the start of the meal.

When the vegetable and protein part of the meal was consumed first, mean postmeal glucose levels were decreased by 28.6%, 36.7%, and 16.8% at 30, 60, and 120 min, respectively, and the incremental area under the curve (iAUC) was 73% lower. Postprandial insulin levels at 60 and 120 min and the insulin iAUC were also significantly lower when protein and vegetables were consumed first.

Meal Order

The authors commented that “the magnitude of the effect of food order on glucose levels is comparable to that observed with pharmacological agents that preferentially target postprandial glucose. Moreover, the reduced insulin excursions observed in this experimental setting suggest that this meal pattern may improve insulin sensitivity.”

How can this be? The same food was eaten, and 15 minutes is not a very long time; it’s a sedate meal, but not a long-drawn out one.

No mechanism is offered in the paper, but we would like to suggest one. After a meal, and especially in people with type 2 diabetes, glucose does not only appear from the food eaten. Glucose is released from glycogen (the liver’s carbohydrate stores) in the fasting state, and this appearance of glucose may continue or even increase after eating carbohydrate, likely because of the post-prandial spike in glucagon, which is released from cells in the gut, as we described in an earlier post. This rise in glucagon in people with type 2 diabetes following carbohydrate consumption has not always been seen experimentally, but it appears as a significant effect in most studies, especially when carbohydrate is fed in a meal; it can clearly be seen here, where the diet is 55% carbohydrate (designed according to AHA and USDA Dietary Guideline recommendations, and represented by the circles – the triangles represent the baseline American diet, which had almost identical effects).
Nuttal 2004 HiCHO glucagon
(from Effect of a High-Protein, Low-Carbohydrate Diet on Blood Glucose Control in People With Type 2 Diabetes. Mary C. Gannon and Frank Q. Nuttall. Diabetes 53:2375, 2004

(This study also says “We and others also have reported that even short-term starvation (hours) results in a dramatic decrease in the blood glucose concentration in people with type 2 diabetes. This seems to be due largely to a rapid, progressive decrease in the rate of glycogenolysis.” Even when you’re fasting, glycogen release slows down once insulin drops and you start burning more fat.)

We suggest that feeding the vegetable and protein part of the meal first blunts the glucagon response to the carbohydrate portion. There could be multiple reasons for this – a delayed appearance of glucose in the gut, or the pre-emptive ability of the insulin, somatostatin 28, and incretins released in response to the protein part of the meal to suppress glucagon release, resulting in lower insulin release overall, because there is both less glucose from glycogenolysis and less glucagon, both of which require extra insulin.
This is only one possible explanation – we’d like to hear others – and verifying or refuting it will take (among other things) accurate measurement of glucagon and other peptides at many time points over the immediate post-prandial period, which is not easy.

Why might eating carbohydrate trigger the release of glucose from glycogen in some circumstances?

Imagine our early human ancestors; they didn’t eat all the time, and sources of dense and easily digestible carbohydrates were rare in some environments. It’s likely that these hominids adapted to hoard glycogen, saving it for an emergency, like sprinting from Pachycrocuta, the giant prehistoric hyena that hunted Homo erectus. And this meant that glucose, which can be used to fuel growth, and to store fat, glucose which we take for granted today, was usually restricted. In this context it made sense if, as soon as an appreciable amount of glucose was detected in the gut, some extra glucose from the glycogen store was released into the blood as well; this extra could be used wisely, and replaced later – the analogy is dipping into your savings for something special when you know there’s going to be lots of money coming in, then being a lot tighter in between paydays.
This (if it’s true) would be the adaptive system, and work when appropriate; and type 2 diabetes, by definition, means that adaptive systems of blood glucose control have become dysregulated and are poorly controlled. It’s a model that does help to explain why fasting and low carb diets can help to bring post-prandial glucose under control, but it’s a highly conjectural one, a “just-so story”.

Takeaways –

  • post-meal release of glucose from glycogen can make an important contribution to post-prandial blood glucose spikes
  • post-meal glycogen release can be triggered by eating carbohydrate
  • very low carb diets and fasting cause the body to conserve glycogen
  •  in this study, eating the carbohydrate portion 15 minutes after the protein and vegetable portion in a high-carbohydrate meal resulted in significantly lower post-prandial glucose and insulin spikes

We note that the carbohydrate timing study of Shukla et al. was “supported by the Clinical and Translational Science Center at Weill Cornell Medical College (UL1 TR000457) and the Dr.Robert C. and Veronica Atkins Curriculum in Metabolic Disease at Weill Cornell Medical College Grant” and that Dr David Ludwig, who was not an author, is thanked for helping to formulate the study hypothesis.

The new BMJ review of high cholesterol and mortality in the elderly


High cholesterol ‘does not cause heart disease’ new research finds, so treating with statins a ‘waste of time’ – or at least that’s the headline in the UK papers following the Just published paper in the BMJ . This, the latest study to question the established view that high cholesterol is always a bad thing, looked at the relationship between serum LDL-cholesterol and mortality in observational studies of people aged 60 and older

“We identified 19 cohort studies including 30 cohorts with a total of 68 094 elderly people, where all-cause mortality was recorded in 28 cohorts and CV mortality in 9 cohorts. Inverse association between all-cause mortality and LDL-C was seen in 16 cohorts (in 14 with statistical significance) representing 92% of the number of participants, where this association was recorded. In the rest, no association was found. In two cohorts, CV mortality was highest in the lowest LDL-C quartile and with statistical significance; in seven cohorts, no association was found.”[1]

In other words, in free-living populations without pre-existing disease, higher LDL-cholesterol predicts a lower risk of death in those over 60, and (with more limited evidence) does not predict an increased risk of death from heart disease or stroke.

This has been known for a long time – this review, which collated these studies for the first time, should not have been controversial. To be more exact, it has been known for a long time that higher levels of total cholesterol are protective in the elderly, and this knowledge predates the use of statins (which, if they are given to elderly with high cholesterol, and do prevent heart attacks, have been proposed as a confounder). Data about LDL-cholesterol levels in the elderly is more recent in most cases, and that is specifically what this review investigates.

The results are not just academic; old people are medicated, and are fed poorly in rest homes and hospitals, on the premise that high cholesterol is bad for them and everything must be done to lower it if it is high. As Professor R.D. Feinman says, “remind me again why we have a medical literature?” – because it’s obviously not there so that anyone can learn from it!

The latest review was published in the UK. Meanwhile the reigning monarch, Queen Elizabeth II, now 90 years old, remains healthy and hard-working on a completely different diet to that recommended and fed to her elderly subjects. She likes her food cooked in butter and cream, eats meat regularly, and skips the carbs.


Her Majesty knows what’s good for her – LCHF!

Criticisms of the study

So, how solid is the latest review? Are there other explanations for the correlations, or lack thereof? The Committee for Evidence Based Medicine launched an attack on Ravnskov et al’s review when it was published, in the form of a post-publication peer review. This is worth reading as it gives insights into possible confounders and errors in interpretation.
Yet it does not really disagree with the evidence or refute or overturn the conclusions.
The main issues are:

Limitations of the study: the search was restricted by being limited to PubMed and English-language publications.

Our response: PubMed is a large data resource, and as can be seen the search turned up many relevant studies. It is not usual to include foreign-language papers in, for example, Cochrane collaboration meta-analyses. The search also retrieved the references from the papers found, a process which would have turned up most of the relevant papers stored outside the PubMed archive.

The limitations of the review are probably related to the limited resources available for research which is merely a matter of public interest.

HDL cholesterol: “only three out of the 19 included studies appear to have controlled for HDL-C level in their analyses. It is highly possible that the observed inverse association for LDL-C and mortality is entirely mediated by a high HDL-C in the included cohorts.”

Our response: the 3 papers that controlled for HDL all found that high LDL-C was significantly associated with reduced mortality.

Statins: “The issue is that those who have high LDL-C are much more likely to be prescribed statins during the period of observation than the low LDL-C group. This would lead to an overall protective effect in the group with high LDL-C, making it appear that LDL-C is correlated with reduced all-cause mortality, when, in fact, it may be the effect of statin therapy.”

Our response: observational studies showing a protective effect of high cholesterol in the elderly date from before the statin era. Statins are expected to reduce cardiovascular mortality; there is no evidence from RCTs that they reduce death from other causes. It is total mortality, not cardiovascular mortality that is most strongly negatively associated with serum LDL-C in this review.  Of the three studies we found in Ravnskov et al. that controlled for statin use, two found a strong protective effect of higher LDL-C and one found no association, a pattern typical of the review as a whole.

Data extraction errors: The reviewers identified a small number of data extraction errors. Unfortunately these are common occurrences; almost all the diet-heart meta-analyses, for example, contain such errors. Hopefully the errors in this review can be corrected. However, using the correct figures does not alter the conclusions of the review.

The CEBM review asks us to reject the conclusions of Ravnskov et al, but fails to supply alternative evidence to support a different conclusion.

More insight:

We pulled up at random one of the studies used in Ravnskov et al’s review to see whether it gives more insight into lipids and risk. This study followed people 50+, so may be of interest to more readers.[2]

Bathum L, Depont Christensenc RD, Engers L et al. Association of lipoprotein levels with mortality in subjects aged 50 + without previous diabetes or cardiovascular disease: A population-based register study. Scandinavian Journal of Primary Health Care. Volume 31, Issue 3, 2013. full-text:  DOI:10.3109/02813432.2013.824157

Objective. This study aimed to investigate the association of lipoprotein and triglyceride levels with all-cause mortality in a population free from diabetes and cardiovascular disease (CVD) at baseline. The European Guidelines on cardiovascular disease prevention state that in general total cholesterol (TC) should be < 5 mmol/L (190 mg/dL) and low-density lipoprotein cholesterol (LDL-C) should be < 3 mmol/L (115 mg/dL).

Design. A population-based register study in the period 1999–2007 including 118 160 subjects aged 50 + without statin use at baseline. All-cause mortality was related to lipoprotein and triglyceride levels and adjusted for statin use after inclusion.

Results. All-cause mortality was lower in the groups with TC or LDL-C above the recommended levels. Compared with subjects with TC < 5 mmol/L, adjusted hazard ratios for the group aged 60–70 years ranged from 0.68 (95% confidence interval (CI) 0.61–0.77) for TC 5–5.99 mmol/L to 0.67 (95% CI 0.59–0.75) for TC 6–7.99 mmol/L and 1.02 (95% CI 0.68–1.53) for TC ≥ 8 mmol/L in males and from 0.57 (95% CI 0.48–0.67) to 0.59 (95% CI 0.50–0.68) and 1.02 (95% CI: 0.77–1.37) in females. For triglycerides, ratios compared with the group < 1 mmol/L in the females aged 60–70 years ranged from 1.04 (95% CI 0.88–1.23) to 1.35 (95% CI 1.10–1.66) and 1.25 (95% CI 1.05–1.48) for triglycerides 1–1.39 mmol/L, 1.4–1.69 mmol/L, and ≥ 1.7 mmol/L, respectively. Statin treatment after inclusion provided a survival benefit without correlation to cholesterol level.

Conclusion. These associations indicate that high lipoprotein levels do not seem to be definitely harmful in the general population. However, high triglyceride levels in females are associated with decreased survival.


In Bathum et al, higher LDL-cholesterol is associated with lower mortality in all age groups over 50.

Our comment: Bathum et al was a study of a Danish population, and the Danes have access to good quality dairy and other animal foods. Thus higher LDL-C, in as far as this has anything to do with diet, may mainly distinguish the well-fed from those subsisting on the ersatz concoctions of the budget food industry. Very high HDL was not protective in “younger” men, but this is something you mainly see in those drinking lots of alcohol. This research does not suggest that the lipid patterns associated with LCHF diets (low TG, high HDL, moderate or high LDL-C) will lead to an increased rate of mortality as people age (which is, of course, when most mortality occurs).

So in the end, we will go with the British headlines – yes High LDL cholesterol isn’t a problem in normal older people – unless it’s low! There are a group of epidemiologists, and nutrition scientists in New Zealand and around the world, who have based their whole career around the idea that high LDL cholesterol is a problem.  What do you have to say?

And, the queen got at least one thing right – low carb.


[1] Ravnskov U, Diamond DM, Hama R. Lack of an association or an inverse association between low-density-lipoprotein cholesterol and mortality in the elderly: a systematic review. BMJ Open 2016;6:e010401. full-text: doi:10.1136/bmjopen-2015-010401

[2] Bathum L, Depont Christensenc RD, Engers L et al. Association of lipoprotein levels with mortality in subjects aged 50 + without previous diabetes or cardiovascular disease: A population-based register study. Scandinavian Journal of Primary Health Care. Volume 31, Issue 3, 2013. full-text:  DOI:10.3109/02813432.2013.824157



UK charity comes out with LCHF Real Food Guidelines

A British charity called the National Obesity Forum, in association with the Public Health Coalition, is in the news today for attacking the low-fat dietary guidelines and food industry manipulation of science (in the UK known as the Eatwell Guide), and suggesting a way out of our growing public health crisis in this document. The Public Health Coalition earlier came out with LCHF guidelines for the prevention and control of obesity and diabetes, and with real food, healthy fat guidelines for the rest of the population. You can download this document, Healthy Eating Guidelines & Weight Loss Advice For The United Kingdom, here.

They’ve even created some cool graphics to counter Public Health England’s dire grain-based Eatwell Guide:


The National Obesity Forum/Public Health Collaboration report has had excellent coverage in the UK media, their recommendations are in all the major newspapers, news websites, and TV news programs.

This has stirred up the usual backlash from experts, many from the UK organizations that have sweetheart deals with the food industry.

For example,

Dr Tedstone of Public Health England responded to the publication by saying: “In the face of all the evidence, calling for people to eat more fat, cut out carbs and ignore calories is irresponsible.”

She said thousands of scientific studies were considered as part of the official guidance adopted throughout the UK, whereas the National Obesity Forum quoted just 43 studies, some of which were comment pieces.

She added: “It’s a risk to the nation’s health when potentially influential voices suggest people should eat a high fat diet, especially saturated fat. Too much saturated fat in the diet increases the risk of raised cholesterol, a route to heart disease and possible death.”

You get the picture. Let’s all get lost in the “totality of the evidence” and leave things as they are, with the Food Industry in charge of our diets in a token partnership with these expert bodies it sponsors.

So who are these irresponsible people at the Public Health Collaboration? They include GP David Unwin, who has published several papers about the effectiveness of LCHF in treating obesity, type 2 diabetes, and non-alcoholic fatty liver disease in his Southport practice. Dr Unwin’s practice has been reported as saving £45,000 per year on drugs for type 2 diabetes compared to the average in his area. (If you want a guide as to how to introduce the LCHF diet into clinical practice in an affordable, practical way, we recommend Dr Unwin’s work)
The PHC also includes cardiologist Aseem Malhotra, who promotes a low carb, high-fat Mediterranean diet as an effective way of managing coronary heart disease, plus a dietitian, a diabetologist and several more GPs. Of particular interest is the presence of a psychiatrist and psychologist, showing a holistic understanding of the importance of not only finding solutions, but motivating and helping people as individuals.


All these people have considerable experience between them, often working in deprived areas and with seriously ill patients; they are reporting back from the front lines in the war on diabesity and the health problems caused by poor nutrition, and it would be foolish not to take them seriously just because they have produced a short, readable document.

The PHC’s Healthy Eating Guidelines & Weight Loss Advice For The United Kingdom joins the ranks of Real Food Guidelines globally – including our own Real Food Guidelines from 2014, as well as the Brazilian dietary guidelines and the Canadian Government report on obesity.

It’s the Real Food Revolution and it’s not going to be won in a day, but when you get a group of well-qualified people leading by example and ignoring the nay-sayers to show that the LCHF approach works, as the members of the PHC are doing in Britain, it’s no longer just a battle of words. The public are impressed by results, rather than by hearing opinions, or by reading thousands of peer-reviewed papers. Which we still do, because reading peer-reviewed papers is necessary, useful, and often rewarding, but by golly it’s nice to see people helped by LCHF to a better life on TV, as in the recent Aussie TV show The Saving Australia Diet.