Well blow me down! Another Paleo Study is Out!

As I mentioned in my last post, there's a new paleo diet comparison study out!  Long term ... two years!  Woo hoo!!  

Hold your horses.

One would think that before taking to social media to tout these things, folks like Robb Wolf, Stefani Ruper and Chris Kresser could hold their tongues long enough to at least obtain and skim the full text of the study they're going to cheer about to try and sell you stuff.  Long-term effects of a Palaeolithic-type diet in obese postmenopausal women: a 2-year randomized trial.  (link is to full text PDF).  

The Study  

This study was done on obese, post-menopausal women, with further characteristics shown in the table below, including some converted/estimated values.  The baseline energy expenditure and intakes are also shown.

Prescribed Diets
  • Randomized assignment to diet
  • Subjects blinded to assigned diet (IOW they weren't told what diet they were being put on by name)
  • Both diets were ad libitum

  • Support:
Each group took part in a total of 12 group sessions held by a trained study dietician (one dietician per diet) throughout the 24-month study period. The group sessions consistedof information on and cooking of the intervention diets, dietary effects on health, behavioral changes and group discussions. The subjects were given recipes and written instructions to facilitate the preparation of meals at home. Eight group sessions (four cooking classes and four follow-up sessions) were held during the first 6 months of the intervention.  Additional group meetings were held at 9, 12, 18 and 24 months.
  • Diet assessment:  
4-day estimated self-reported food records conducted at baseline (2 4 days) and monthly until 6 months, thereafter at 9, 12, 18 and 24 months. Subjects were instructed to keep a record of all food items consumed over 4 consecutive days (three weekdays and one weekend day) and to describe and estimate the amount of food eaten by using coloured food-portion photographs representing known weights and household measuring utensils (for example, cup, spoon and standard weight of food items).

Actual Diets

values in ( ) are negative meaning falling short of target or differential

This is a rather big deal, because what is the purpose of this study, if not to compare the "long term" effects and effectiveness of two dietary regimes? It's not like they even came close either!   At the very least, the so-called "controversial" aspects of the diet should be adhered to so they can be tested.  
  • The paleo diet is thought by some to be too high in protein.  The PD group didn't even hit 25% let alone the 30% protein they were supposed to reach.  (This was verified by a urine test).
  • Critics of low fat diets say that 25-30% is too low in fat.  The NNR group barely lowered fat at 6 months, and actually increased it ever so slightly percentage-wise from baseline at 2 years.  So as has been seen in the past, so-called low fat diets generally don't alter this component much.  But they don't even meet the upper range of 30%!  (My values use the average 27.5%)
  • Critics of high carb diets say that 55-60% of intake is too much.  The NNR group not only failed to meet this goal as any sort of a test of this claim, but they actually went the other way!    That's right.  In absolute terms the NNR group cut carbs  from baseline intake!  Their reduction in carbs, in absolute grams, was about 40% of the carb reduction of the PD group.
Values in ( ) are negative, Fiber and Evil Sucrose are in grams
  • In terms of one of the only two emphasized food types, the NNR group does not appear to have paid attention to the fiber content of foods.  While they likely ate a bit more fibery foods, the absolute amount of fiber in their diet essentially flatlined.  
The above render this study effectively useless for the purposes for which it was undertaken.  How can you ascertain anything meaningful when you aren't even studying the intended treatment?    This is worse than studying a 100 mg daily dose of Drug A by administering a 75 mg equivalent dose of Drug B. 

We are told that the dietary assessments were done monthly for the first 6 months.   What is the point of continuing this study for two years, or even the six months, if after 1 month it was clear that neither group was even adhering to the macro recommendations?  Why, in those cooking classes and counseling sessions were they not counseled to, for example, eat more lean meat (for protein in the PD group) or eat more fruit (for carbs and fiber in the NNR group)?  Why?  What were the investigators thinking???

There Was Attrition

On the plus side, they only analyzed those that stuck with the diets instead of the awful intent-to-treat analysis.  Attrition was high, with 8/35 = 23% already lost at 6 months in the NNR group where only 22/35 = 63% completed.  For PD, it was a bit better losing only 1/35 = 3% at 6 months, while 27/35 = 77% completed.  Just how this attrition altered the results here is unknown.

The Results

There really are only two statistically significant results.  The "PD"-like diet outperforms the "NNR"-like diet at 6 months with weight loss of 6.5 kg (14.3 lbs) vs 2.6 kg (5.7 lbs) being statistically significant.  This translates to other weight-related parameters. Here are the body weight and measurement result plots.

Now, the PD still seems "ahead" at 2 years, but NS = Not Significant (statistically).  Just in case you're inclined to see this as still meaningful, I crunched a few things and shifted the NNR curves to match PD at 6 months.  Still NS, but over the long term, the PD trend vs. NNR was neither impressive nor better by trend.

The only other statistically meaningful difference of all that was measured, was the fasting triglyceride levels for the PD group. There was a drop from 108 to 74 mg/dL at 6 months, while the levels rose slightly for a net reduction from 108 to 88 mg/dL at 2 years. I would remind the audience that under 150 is "normal" despite what certain low carb advocates seem to believe.  Basically, the triglycerides were the ONLY characteristic that differed between the groups in the long term, and going from normal to a little lower normal is quite the bust here if you ask me. 

Some Basically Meaningless Banter
  1. In absolute amounts, the NNR group reduced evil sugar intake by more than the paleo group. Both groups still certifiably not addicted at 2 years.
  2. Both groups ended up decreasing absolute fiber intake slightly on their "healthier" diets. 
  3. While the PD group increased O3s compared with NNR, they also increased O6s.
  4. The PD group actually reduced saturated fats more than the NNR group on both an absolute level and as a percentage of total energy. The NNR group actually consumed about 1 teaspoon more of pure sat fat at 6 months, and around 1/2 tablespoon of more at 2 years. 
  5. The most significant change seems to be in the MUFA consumption where the PD group was consuming about 1 tablespoon more MUFA than the NNR group at 6 months, but only about 1 teaspoon at 2 years. 
  6. The PD group dropped sugar less and sat fat more and had lower triglycerides. Make of that what you will ;-)
  7. Apparently the PD causes increased food intake amnesia with time as they under-reported intake more severely by the two year mark while gaining weight.  Either that, or as in the LC arm in Shai, we are to believe that they cut intake further during the latter phase yet regained some weight.

The "Paleo" Diet 

I will keep beating this drum.  Neither the "Paleo Diet" prescribed in this study, nor the one actually implementd, bears any resemblance to the popular implementation of the diet put forth currently by Wolf, Sanfilippo, Sisson, Gedgaudas, Jaminet or Kresser, to name a few, and especially not Dr. NBA Shanahan!  Not a one of them.   There is:

  • NO dairy - not even a pat of Kerrygold let alone sticks, no parmesan, no mascarpone, no whipped cream, no ghee, no kefir, no yogurt.
  • REDUCED sat fat (reduced more on an absolute level in the PD group than in the "mainstream" low fat NNR).  No tallow or lard.
  • NO bacon or other fatty meats
  • INCREASED MUFA and PUFA (including O6)
  • CANOLA oil allowed and used
  • NO coconut oil
  • NO chocolate
  • NO ketosis
  • NO white rice
  • NO kombucha
  • NO bone broth
  • NO tequila
  • NO coconut and almond flour banana bread caveman one blender wonders
  • .... you get the point.
This diet, either prescribed or implemented, is far different from that put forth by Cordain and Wolf in their fad diet books, however obsolete we are supposed to believe those books are.  

Therefore ....  If ANY of these gurus want to use this study in support of their advocacy of their increasingly nebulously defined and contradiction-riddled "paleo" diet, I'm going to call them on it, and I hope you will too.  It is DISHONEST.  Period.

To Be Continued ...

I had intended to do a "there are some meaningful takeaways" section here.  I had a snow day and some unexpected time here.   But alas, I've run out of steam and this has gotten long enough already.  There are some lessons to be learned from this for us postmenopausal women (and the people who love them ;-) ), and just women in general, vis a vis current diet trends and bogeymen.  I promise to follow up as soon as I can.

😇😈 Snark Time 😈😇

Minor edits were made Feb 1, 2019.  When originally posted in February of 2014, I led with this snarky bit about Robb Wolf's classy announcement of the new paleo study that will quiet the nay-sayers.  Still waiting on that major paper on paleo and Autoimmunity ....

The police chief in Reno and the powers that be at Specialty Health must be so proud to have such an endearing cockalorum as ambassador for their nothing-really-all-that-special-or-different CVD-risk-mitigation-through-low-sat-fat-diet-weight-loss-and-conventional-statin-and-other-meds program.   

This is surely the way to court critics and gain converts.  


charles grashow said…
Robb Wolf The
nay-sayers will (rightly) point out that any whole foods, protein
centric approach is going to produce similar WEIGHT LOSS patterns.
Check. What these other approaches will NOT address is autoimmunity and
the plethora of systemic inflammatory issues we see. So, please take
each of these in context while also remembering the whole picture. As to
the protein piece: folks found it hard to eat as much as they were
supposed to. Think on that for a moment.
Jane Smith said…
Also check this nice recipes www.paleo99diet.com
silentshade12 . said…
Hehe. These guys are such a joke. It's a bit off topic but I stumbled upon some papers on the rice diet ( I had never heard of it before) which has been used to treat obesity in the past. The diet consist only of rice, fruits and fruit juice and there was massive weight loss and some amelioration of metabolic parameters in obese patient.

I'm still wondering how the LC gurus would explain away those results.


charles grashow said…

charles grashow said…
1) Before another study is done using a paleo diet doesn't there have to be consensus on exactly what the diet is?

2) With regard to IF among paleo dieters let's go to Cordain

3) With regard to paleo proponents ignoring increases in LDL-C again let's go to Cordain


4) With regard to the components of a paleo type diet - again let's go to Cordain

Take a look at Table 2

Consume a diet high
in fruits, vegetables, nuts and berries and low in refined grains and sugars.

Nutrient dense, low
glycemic load fruits and vegetables such as berries, plums, citrus fruits,
apples, cantaloupe, spinach, tomatoes, broccoli, cauliflower, and avocados are

Increase consumption
of omega-3 fatty acids from fish, fish oil, and plant sources.

Increase consumption
of lean protein, such as skinless poultry, fish, and game meats and lean cuts
of red meat. Avoid high fat dairy…

SO - until there is some kind of consensus aren't these types of studies useless?
eulerandothers said…
Did you notice this in the related citations on the NCBI page for that study?


Women in both the dark chocolate snack and non-chocolate snack groups, respectively, experienced decreases (P<0.001) in body weight (-5.1 vs -5.1 kg), hip circumference (-5.8 vs -5.4 cm), waist circumference (-5.7 vs -3.5 cm), fat mass (-3.9 vs -3.6 kg), and body fat percentage (-3.4% vs -3.1%), with no change in lean mass. Improvements in anthropometric and body composition measurements among overweight and obese premenopausal women can be achieved with a reduced-calorie diet including either a daily dark chocolate snack or non-chocolate snack.

Call it the 'Chocolate Diet' and publish a book about it. The chocolate bars were Hershey Bars ('Big Candy') so you know the results are tainted.
eulerandothers said…
I know someone who went on the Rice Diet at their campus in North Carolina years ago. She lost a lot of weight but eventually gained it all back.
LC gurus don't have to explain away those results. They don't have to explain away anything. Who sez they gotta?
charles grashow said…
By his definition of the paleo diet Robb is probably in full compliance
silentshade12 . said…
Well, when you make claims such as that carbs are inherently fattening and that sugar increase CHD risks and that calories do not matter as long as you do not eat carbs but then we see scientific reports of people losing a lot of weight and seeing biomarkers improve (althought weigth loss usually improve CHD risks markers) on a starch and sugar diet, you're kind of losing face aren't you?

Talk about a big black swan that disproves their conjectures.

But you're right. They would have too much to explain away anyway so it's better to put your head in the sand and act is all of this does not exist.
charles grashow said…
charles grashow said…

Study protocol

A healthy diet with and without cereal grains and dairy products in patients with type 2 diabetes: study protocol for a random-order cross-over pilot study - Alimentation and Diabetes in Lanzarote -ADILAN.


We intend to include 15 adult patients with a medical diagnosis of type 2 diabetes
mellitus with or without medication and with an increased waist circumference (≥ 80 cm
for women and ≥ 94 cm for men) in a random-order cross-over diet intervention study
during two periods of four-weeks separated by a six-week washout period. Patients
will be instructed to eat two healthy diets according to official dietary guidelines
with respect to macro/micronutrient composition and fiber content, but differing in
the type of food included, with one diet being without cereal grains and dairy products.
Lunch will be served in a hospital kitchen for control of nutrient intake, while the
rest of the meals will be eaten at home according to specific directions. The energy
content of the diets will be individually adjusted to maintain a stable body weight
during the two four-week intervention periods. Primary outcomes will be change in
fasting plasma glucagon and fructosamine, while secondary outcomes include change
in fasting glucose and glycated hemoglobin, glucose and glucagon response during oral
glucose tolerance test, blood lipids, blood pressure, C-reactive protein, body composition,
quality of life, subjective experience with the two diets, satiety scores and changes
in medication.
charles grashow said…

Beneficial effects of a Paleolithic diet on cardiovascular risk factors in type 2 diabetes: a randomized cross-over pilot study

There are virtually NO DIFFERENCES between the two groups after 6 months

The paleo diet group ate a lot MORE fruit and ate LESS saturated fat

How is this a paleo diet according to the "experts"?
carbsane said…
I've spoken to enough of them and followed enough AHS/PFX twitter feeds to know few live by what they preach. In Robb's case, it's whatever makes him feel good and what he can't live without. So basically gluten is out, anything else goes. But I don't care so much about that as I do about trying to tout these studies as support for the diet he promotes. That is false advertising.
carbsane said…
We eat a lot of protein as a culture. That Bray overfeeding study demonstrated that you gain a shit ton more weight for the same calorie surplus on high protein because of gaining lean mass. Some of this lean mass does NOT need to be preserved and there's as much room for such a diet as there is for the keto nonsense!
charles grashow said…
Jaie Jac said…
"BLOW ME" real mature, robb.
charles grashow said…
Why do you refer to the Perfect Health diet as "nightmarish"?
Jaie Jac said…
White rice and excessive Saturated fat is Unhealthy, IMO.
Screennamerequired said…
Exactly. They complain about how badly saturated fat has been unfairly vilified but their paleo diet is even lower in saturated fat then the NNR diet.
Go to a paleo forum and post that your following this diet and ask for feedback. I guarantee you'll be told to add more fat from coconut oil or ghee.
charles grashow said…
the original paleo diet did contain saturated fat.
charles grashow said…
So sayeth the vegan. Here's the deal - I don't think that saturated fats are unhealthy per se - so for every paper you produce that says it is I can probably find one that says it isn't.

Which will prove nothing
Correction. 'White rice and *excessive* saturated fat' sayeth the Vegan.
Jaie Jac said…
I am not a vegan.
silentshade12 . said…
Charles, they raise cholesterol, they are inflammatory, they induced insulin resistance. SFAs effects on health is a net negative. They should be kept to a strict minimum.


I mean the idea that SFAs are detrimental is not even up for debate IMO. I'd gladly look at paper showing that SFAs improve markers of CHD risks.
silentshade12 . said…
I'm not sure i'm following you here. Are you saying that these guys were heavy because they were eating a high-protein diet and then they lost a lot of weight because a lot of it was muscle mass?

I'm not endorsing that diet btw. I was only using it as a blackswan to show that you can lose weight on a starch and sugar diet.
silentshade12 . said…
Why didn't my comment to Charles made it trought moderation? Are you censoring differing viewpoints?
silentshade12 . said…
I don't understand why I can't answer to Charles in regard to SFAs. Is there a problem with the links i'm providing?
charles grashow said…
The why do you say that "White rice and excessive Saturated fat is Unhealthy."

1) Define what you mean by excessive saturated fat.
Jaie Jac said…
by excessive, i mean far exceeding general guidelines by organizations such as the American heart Association. some within "paleo" go full rain man and eat sticks of butter in their coffee. that's excessive. Having a piece of cheesecake every single day in moderation is fine, just as long as your entire diet consists of healthier fats, fiber, vegetables, etc.
charles grashow said…
We are in agreement here
Screennamerequired said…
charles grashow said…
charles grashow said…
Jimmy Moore
6 hours ago
so here's the verdict on my cholesterol test results from Sam's Club
today: My total cholesterol was 392 and my HDL was 70. When the
phlebotomist saw my total reading, she immediately responded "OH MY!"

I asked what the HDL was and she shared it was 70. "Good," I said. She
immediately asked, "But did you see what your total cholesterol was?" I
retorted, "Yes but I'm not worried about it." If I could have heard what
she was saying inside of her head, I'm sure it wouldn't have been

Always fun to add a little shock and pizzazz to the cholesterol screeners at Sam's Club when they do these health checks.

SO - he sees NO problem with a TC of 392 just because his HDL is 70!! This man is delusional. Also - the didn't test for LDL or Trigs.

My questions are
1) will he take a NMR test for his new book
2) What does his co author - Dr. Eric Westman - have to say about his lipid panel?
Jaie Jac said…
exactly! Just the other day on a "paleo" Q/A forum, i saw an individual who is very active, yet follows the high fat, high sat fat paleo diet. Upon presenting their worsening lipid panels on the site, noticeably confused as to why it got worse, the other posters only repeated the same "avoid the carbs, or particle size matters" mantra and totally ignored the elephant in the room. The Echo chamber is often loud on these type of sites.
Sanjeev Sharma said…
> Again - please tell me why this is NOT a problem.


Also, I find your lack of faith ... (James Earl Jones voice) ...

Jaie Jac said…
Would it be safe to say that jimmy is effectively dying? that plaque must be building rapidly. I hope he's right- otherwise he is only marching towards a cardiac event.
Sanjeev Sharma said…
if the specific pages are being blocked are you sure they're safe (free of malware)?

fudge up a google search that produces the pages you want and copy & paste the google url

it looks like this:

Screennamerequired said…
"Again - please tell me why this is NOT a problem."

Cholesterol is essential for life!
charles grashow said…
I agree that cholesterol is essential. Our bodies produce between 800-1100 milligrams per day - so just how much dietary cholesterol do we really need to consume?
silentshade12 . said…
They were pubmed link to studies... i doubt they are problematic.
carbsane said…
Yeah -- it's a possibility that some weight came from lean, and moreso with low protein. Not saying anything definitive. This is a great example you found for your intended purpose. In the end you have caloric deficit = weight loss, but if low protein break down lean mass which is less calorically dense thus break down more.
carbsane said…
Just another random FYI -- Discus dashboard again on the fritz so I am unable to approve anything that might be in the hopper.
charles grashow said…
So - on his last test JM's TC went up to 392 and his HDL was 70 which indicates a LDL of >300.


TC - Sure, it spiked up above 400 late last year, but keep in mind I had lost around 55 pounds at that point and weight loss can indeed raise your cholesterol. But now that I’m returning to a more stable weight, my total cholesterol has dropped over 100 points since December 2012. AMAZING! How many people take a statin drug to lower their total
cholesterol when perhaps their body would have naturally taken care of that without the medication? It just breaks my heart! Yes, this level is still showing there is something else perhaps going on and I’m exploring those issues personally. But it doesn’t mean I need to
automatically jump to statin therapy.

LDL - Check out my latest LDL-C at 236. While that is still considered very high by conventional wisdom standards, it also reflects my cholesterol numbers becoming more normalized again after soaring up above 300 last year. Perhaps the trend will continue to go down as my weight stabilizes even more. In fact, that’s the lowest level of LDL-C I have seen since 2009. While I don’t put much stock in LDL-C meaning anything that warrants treatment, it is an interesting number to watch for determining if something else is going on.

LDL-P While there is debate about whether it is the LDL particle number or LDL particle size, one thing is for sure–you don’t want a lot of Small LDL-P in your body! As you can see, my high-fat, moderate protein, low-carb approach has kept this number right where it needs to be. Ideally you’d like for it to be as close to zero as possible, but I’ll take what I have. I will continue to monitor this to make sure I stay under 500 over the long term. It will be interesting to monitor this as my LDL-P keeps dropping as I predict it will.

The man is in TOTAL DENIAL - his lipid profile is getting worse and he see no problem.

BTW - with regard to his CACS score of zero


Of 118 patients, 89 (75%) had a CACS of zero and 94 (80%) were men; mean ± SD age was 48.9±5.7 years. The mean FRS of this group was 4.0; 86 patients (97%) were considered at low risk (<1% annualized rate) of cardiovascular events. Evidence of carotid atherosclerosis was found in 42 (47%; 95% confidence interval, 37%-58%) of these 89 patients; carotid plaque was found in 30 (34%); and CIMT above the 75th
percentile was found in 12 (13%) of age-, sex-, and race-matched control patients.

Subclinical vascular disease can be detected by CIMT evaluation in young to middle-aged patients with a low FRS and a CACS of zero.

SO - has JM had a CIMT test done??
charles grashow said…
Jacob Gudiol said…
I've had personal contact with the first author of this study several times. We had a couple of email-exchanges after this study was published and she doesn't not believe that the results comes from the paleo-diet but from greater restriction. She also thinks that participants in the paleo group underreported their carb intake. A lot of the participants in the NNR group also felt like they were in the placebo group.

They have also measured a lot more in this study and there will be more studies coming out about compliance, liver fat, fat biopsies, hormonal measurements, psychological stuff etc...

The actually measured compliance to the paleodiet by measuring 15:0 and 17:0. I'm not sure but I believe they could have included a test for Alkylresorcinol as well.
Unknown said…
Tools for Applying the Event Study Methodology and News Analytics: Abnormal Return Calculator,Text Analysis,Regular Expressions,Data Sources - http://www.eventstudytools.com
carbsane said…
Sorry this comment slipped by me! Thanks for sharing this. The other stuff will be interesting, perhaps moreso than this. Too bad there wasn't really diet compliance to compare between the two diets.
Blogger said…
EasyHits4U - Your Free Manual Traffic Exchange - 1:1 Manual Exchange Ratio, Business networking. Get FREE Advertising!
Blogger said…
Having Problems STICKING with the Paleo Diet?

Do you want to munch on healthy delicious recipes TONIGHT?

Check out: Paleo Hacks Cookbook.
Blogger said…
New Diet Taps into Pioneering Plan to Help Dieters Lose 20 Pounds within Only 21 Days!