I recently listened to a very interesting conversation on the Doctor’s Farmacy podcast with Dr. Uma Naidoo about nutritional psychiatry. I actually found this discussion interesting enough to buy Dr. Naidoo’s new book, This Is Your Brain on Food: An Indispensable Guide to the Surprising Foods that Fight Depression, Anxiety, PTSD, OCD, ADHD, and More. Unfortunately, after reading the book, I now wish I hadn’t wasted my time or money.
When I was about an hundred pages into the book, I decided I’d only finish the book to write a review to share on Amazon. At that point, an hundred pages in, I was going to give the book 3 stars. But the more I read of the book, the fewer stars I decided to give it.
I actually broadly agree with many of the points that Dr. Naidoo argues for in the book including that one’s diet directly affects one’s mental health, there’s a connection between gut health and brain health, and modern Western or standard American diets aren’t healthy. Though, we don’t fully agree on what makes a Western diet unhealthy. Psychiatry today also tends to look at and try to ameliorate symptoms, rather than find and address underlying causes. So, I actually commend Dr. Naidoo for exploring diet and its impact on the gut, brain and mental health as an underlying factor affecting mental health. Poor diets, sleep disorders, malocclusions, interrupted circadian rhythms, environmental toxins, trauma, etc all impact one’s brain, gut health and hormones. Sadly, many of these factors aren’t fully explored before psychiatrists prescribe pills in our drug happy system of medicine.
Unfortunately too, whatever broad general agreement I had with Dr. Naidoo is severely undermined by her reliance on very weak confounded epidemiological and rodent studies as well as her repeated inconsistent nutritional advice including her inconsistent saturated fat phobia, her misrepresentation of Mediterranean diets, and a few other cringe worthy things that I’ll quickly touched on below. Plus on top of that, some of the misrepresentation and or lack of critical analysis of the research cited was downright disturbing. Why? She either didn’t understand what she read or was dishonest about what she read to further her specific dietary biases.
Let’s start with epidemiological studies, where she hyped low relative risk numbers enough times to make me wonder whether or not she knows the difference between relative and absolute risk. As I’ve noted in prior blog posts here and here, stating that there’s a 20 percent (relative risk) increase between the subject and control groups in a study doesn’t mean much of anything without knowing the absolute risk. Such small R/R’s typically are so confounded as to not really demonstrate anything. Small R/R’s also often have absolute risks that are very small in the less than one or two percent range. I think she’s spent too much time at Harvard where fear mongering with very weak and confounded R/R’s seems to be the modus operandi. As the illustration immediately below demonstrates, correlations and associations do NOT necessarily equal causation.
Dr. Naiboo also repeatedly cites rodent studies to typically berate high saturated fat diets. But she doesn’t seem to be aware that the high fat rodent chow diets usually consist of soybean or corn oil and sugar. These diets aren’t natural diets for rodents to consume. Plus such rodent chow diets are quite a bit different than the healthy fats (including saturated fats) eaten in the context of LCHF, keto, paleo, or similar diets. So holding out rodent studies- without breaking down what the rodents actually ate- to prove high saturated fat diets are bad is either naive or dishonest. For example in chapter six, on dementia and brain fog, in back to back studies she cites rodent studies as examples of harmful high saturated fat diets in footnotes 16 (Menay et al 2010) and 17 (Wu et al 2014). In Menay et al 2010, the rodent chow for the high fat diet is specifically noted. This rodent chow #D12266B consists mainly of sugar and corn oil (over 75% corn oil with the remaining fat from butter, a saturated fat). The author falsely describes this rodent chow as primarily a “high saturated fat diet”. In Wu et al 2014, the specific chow isn’t noted. However the high fat diet is specifically described as being high in both saturated (SFA) and monounsaturated (MUFA) fat made from lard and corn oil. Lard from factory pigs is a mix of approximately 40% SFA, 50% MUFA (oleic acid) and 10% poly-unsaturated fat (PUFA). Corn oil is nearly 60% linoleic acid (an Omega 6 PUFA) and 28% oleic acid (MUFA). So is the oxidative stress and other maladies, that Dr. Naidoo attributes solely to the “bad” saturated fat, caused by the SFA or the other “healthy” fats (and sugars) in the rodent chow in this study? Due to how the study was designed including what the rodents were fed, there’s no definitive answer to this question because of the confounders…. despite Dr. Naidoo’s claims to the contrary.
In general, with saturated fats, Dr. Naidoo is kind of silly. To her, saturated fats are “bad fats”. However, she recommends avocados, avocado oil, and coconut oil as “healthy fats”. The most common avocado is the Haas avocado, which consist of 25 to 30% palmitic fatty acid (a saturated fat) and around 30 to 35% total saturated fatty acids. (Some other kinds of less common avocados, like Pinkertons, have less saturated fat- around 20% SFA’s). Coconut oil is largely lauric, capric and palmitic SFA’s. So coconut oil is over 80% SFA’s. But similar fatty acid compositions in red meats per her repeated assertions are “bad unhealthy fat”. She doesn’t seem to understand that most sources of fat consist of different ratios of saturated, mono-unsaturated and poly-unsaturated fatty acids. I’ve seen pastured lard as high as 60% oleic fatty acid, a mono-unsaturated fatty acid (olive oil is around 70 to 75% oleic fatty acid). Pastured tallow is also primarily monounsaturated/poly-unsaturated fatty acids with guess what saturated fatty acid? Yep, you betcha…. palmitic fatty acid. Guess she didn’t take any lipidology courses when she got her nutritional degree.
But that’s not the only inconsistency regarding red meat. She also notes not to eat grain finished red meat because of its high Omega 6 to 3 ratios (around 15 to 1) yet writes to eat “healthy fat” from almonds as well as eat almonds in general. Hmmm….all beef (whether grain or grass finished) has fairly low amounts of omega 6’s and 3’s, so it’s not a good or bad source for either. But almonds, on the other hand, are really high in Omega 6’s and their ratio of 6’s to 3’s is around 2000 to 1. Many nuts and seeds are high in Omega 6’s and very low in 3’s including Brazil nuts (1000 to 1), and pumpkin seeds (175 to 1).
Funny too she advises against eating wheat bran because it’s high in phytates (phytic acid) that will block mineral absorption. She makes this recommendation almost immediately after suggesting to eat other foods like pumpkin seeds and Brazil nuts… that are guess what? Yep you betcha again, very high in phytates that block mineral absorption including zinc, selenium and iron. Brazil nuts are one of the highest nut sources of phytic acid. (Vegans are routinely deficient in zinc and iron. It’s not from not eating enough pumpkin seeds. The zinc, and iron in nuts, seeds and many plants just aren’t very bio-available since phytates – as well as oxalates- are chelators that bind minerals).
But the many inconsistencies don’t end there…..She advises people not to eat foods that easily oxidize, but suggests people should eat canola oil instead of soybean oil due to the Omega 6 to 3 ratios. Well guess what? When canola oil is expeller pressed and hexane extracted, it’s oxidized so much during production that it has to be deodorized to mask the rancidity (see video below). Some of her recipes also use canola for cooking. When polyunsaturated fats are heated during the extraction of the oil or during cooking, the bonds break and the oils form plant sterol oxidation product [POPS]. These POPS effect membrane function and cause inflammation, thus they have been connected to arteriosclerosis (Vanmierlo et al, 2012). Furthermore with canola oil, she also mentions a “Norway Diet” that uses canola oil instead of olive oil. Canola oil wasn’t even created until the 1970’s so there never was any Norwegian diet that ever relied on canola oil as its primary fat source.
There also was no single Mediterranean Diet. Whether you’re in Italy, Spain, France, Turkey, Greece, etc, there are multiple Mediterranean diets many of which include a lot of cured meats (salumi, charcuterie) , fermented dairy (cheese, yogurt), pork, lamb and fish. These are not “plant based” diets. France eats a ton of butter as well yet somehow doesn’t have high heart disease or more mental health disorders than other European countries. The “Mediterranean Diet” that the author describes isn’t what’s consumed in this region. As Dr. Tim Noakes explains the “Mediterranean Diet” the author describes is something Walter Willett made up at a conference sponsored by the olive oil industry in 1993.
Too often many of the claims she made regarding the benefits or detriments of different foods based on “this study” or that study were a bit incredulous. So, frequently I found I had to refer to the footnotes to find the studies she was using to support her claims and then review those studies for myself. After reading through many of these studies, it was very apparent that she was frequently overstating what was demonstrated by the actual studies. This occurred enough times that I had to pretty much question the veracity of any and every claim she made. For example, in chapter 10 on libido in footnote 13 (Tremellen 2016) , she makes a statement, cites a theory (hypothesis) noting the theory “purports” that the diet caused dysbiosis (leaky gut) which, in turn, caused the sexual dysfunction. Dr. Naidoo then concludes that this hypothesis demonstrates a causal connection of gut health to sexual health. The study clearly notes it’s a hypothesis based on epidemiological studies that can’t show causation. So even though there may indeed be a connection between gut health and sexual function, the specific study cited doesn’t provide anything more that a hypothesis based upon circumstantial evidence.
Moreover one really has to question her analytical abilities when she reads a study. Why? Since with any study one reads, one needs to understand the methodologies and review the raw data in order to critically assess the conclusions rather than just cherry pick whatever language supports the author or researcher’s biases. For instance in “another study” (St-Onge et al 2016) in chapter eight on insomnia and fatigue, Dr. Naidoo notes high saturated fat, high sugar and low fiber diets provide less restorative sleep. This study had only 27 participants (14 men and 13 women), who were very rigorously screened so as to not have any sleep disorders or other anomalies that would adversely impact their sleep. The participants were randomly assigned to two groups for restricted and habitual sleep. They were tested in a crossover design for these two different sleep patterns for six days each. During the first four days of each pattern they were given specific meals at specific times. During the last two days of each pattern, they could eat whatever they wanted to at any time. This was their ad libitum diet. They were sleep tested on the third and fifth day of each sleep pattern based on the different diets. An analysis of diets for the ad libitum diet showed that on average there was a very slight increase of saturated fat intake from 7.5% to 10% , a slight decrease in protein and a slight increase of carbs, though less fiber. Per the sleep studies, these slightly altered ad libitum diets resulted in less slow wave sleep [SWS] and longer sleep onset latency [SOL]. No data is given as to when meals were consumed during night 5 and 6. So basically all of these associations are confounded not only by what was consumed , but when whatever ad libitum was consumed. Was it the fat, protein, fiber, a micro-nutrient or the time a meal was consumed that resulted in the lower SWS and longer SOL? Don’t really know and can’t really tell from how the study was set-up. Why did the researchers not control meal times or provide pre-determined diets with fewer variables on nights 5 and 6? Don’t know, but the results end up being confounded and the associations are thus very weak. Furthermore without looking at the raw data, it’s hard to determine whether or not an outlier or two skewed the numbers in such a small sample group.
Anyway, I could go on about a number of other items in the book like not differentiating between vitamins A1 and A2 or giving bogus plant sources of B12 or not realizing that Glycemic Indexes [GI] are averages where different people have very different glycemic responses to the same foods in large part due to their different microbiomes (Zeevi et al 2015). I’m sort of surprise she didn’t realize this GI issue since the book is supposedly about the brain-gut connection. Like our brain, ketones can also be used as an alternative source for the gut. So she also doesn’t seem to understand that the gut lining can also get isobutyrate from protein fermentation and from ketones via a blood pathway (see video below) instead of butyrate from the fermentation of microbe accessible carbohydrates (MACS a.k.a. fiber). Thus eating dietary fiber isn’t the only way to feed the gut lining short chained fatty acids. Humans evolved in places with wet and dry seasons as well in places with winters where plant foods weren’t readily or seasonally available. Thus we’re metabolically flexible. Not adhering to seasonable dietary patterns may have actually SIMPLIFIED modern human gut microbiomes.
So, in short, I agree that modern Western diets are bad for mental health…but probably more so for easily oxidized industrial plant oils and excessive amounts of sugar rather than anything to do with the saturated fats in red meat. Red meat, which Dr. Naidoo routinely labels as “bad”, has been part of the diets of the homo genus for over two and a half million years. So red meat isn’t unique or new to “western diets”. Though industrial plant oils like canola, corn and soybean oil are and have only been consumed for the past one hundred or so years in modern diets. Additionally, deep frying in these oils generates hydroxynonenal. These lipid peroxidation products may play a role in Alzheimer’s and other lifestyle diseases (Yamashima et al 2020). Mutagenic forms of wheat, high fructose corn syrup and glyphosate are other novel new foods, ingredients or residues new to our species’s diet that are omnipresent in modern Western or standard American diets. These items, like easily oxidized highly processed industrial plant oils, all adversely affect gut and brain health.
In many of Dr. Naidoo’s case studies from her own practice, her patients go from eating junk food from fast food restaurants full of these deleterious items listed just above to eating whole and better prepared food. So is it really any wonder that her patient’s mental health improves? Take Letitia in chapter four on trauma for example. Letitia’s diet consisted of Chik-fil-A deep fried chicken sandwiches, french fries and soda. What’s used in deep fryers? It’s not beef tallow. In quick serve (and most) restaurants the fryer oil used is almost universally soybean oil. (Chik-fil-A uses a pressure fryer). This soybean oil is heated and reheated numerous times plus filtered to extend its life. This is highly oxidized (rancid) polyunsaturated fats full of free radicals and other toxic end products. The sandwich and fries usually are also consumed with condiments- like ketchup- full of high fructose corn syrup. Soda is full of HFCS and other food colors. The artificial sweeteners in diet soda are also bad. Yet somehow per Dr. Naidoo’s repeated analysis, red meat is the repeated villain in her analysis with its “bad saturated fat”.
Her anti-red meat attitudes make me wonder whether or not the author is Hindu? Or maybe she’s just spent too much time around Walter Willett and thus thinks “adjusted” epidemiology proves causation. I wonder because her incessant haranguing of red meat and her reliance on weak confounded epi-studies borders more on religious belief than scientific method. Sadly, association and correlation are probably Dr. Naidoo’s most frequently used words in the book. The reality is that epidemiological studies are intended to derive hypothesis for further more controlled research. Unfortunately, in nutritional science, that controlled research is very difficult and expensive to do. So, that research often doesn’t get done. Plus it’s worth noting that when epidemiological research is followed-up by more controlled clinical research, 80% of the time the hypothesis derived from the epidemiological study is demonstrated to be false (Stanley & Karr, 2011).
The Doctor’s Farmacy podcast discussion with the author was very good. So, I’m pretty bummed that this book is such a mess. Obviously simply eating better, including not eating junk food fried in industrial plant oils, will have a beneficial impact on one’s gut and brain. So to a degree the book provides some value, especially to people who know less about nutrition than the author (including her patients). Though as demonstrated above with just a few examples, there’s a poor understanding of and over reliance on very weak confounded science. So many of the conclusions made are not as definitive as the author asserts. Hopefully, the author was simply unaware of her poor understanding of this science rather than being purposely dishonest. To that end, I’ll give her the benefit of the doubt. Though the book only gets one out of five stars.