Aug 19 2010

Alzheimer’s and Diet

The pattern is depressingly common – take a disease that is not currently cured by science-based medicine. Claim that doctors “don’t care”, or are ignoring treatments that do not make them money, or there is a conspiracy of silence headed by “Big Pharma.” Then offer a snake-oil alternative based upon anecdotal evidence. Sometimes faith and God are thrown in for good measure.

Nita Scoggan fits this mold nicely. She is promoting herself as a “health and happiness coach” and using her husband as anecdotal evidence of the power of her nutritional advice. The story is now being promoted by other low-carb gurus, including Jimmy Moore.

The story is a great example of why anecdotes are so problematic. We are told that Nita’s husband, Bill, was diagnosed with Alzheimer’s disease, but she cured him with diet and nutrition, specifically coconut oil and a low carb diet. The problems with this story begin with the diagnosis. Scientific medical trials always include what are called inclusionary and exclusionary criteria – this is to make sure that subjects actually have the disease that is being studied. This is critical because apparent remissions may be due to the fact that the person never had the alleged disease in the first place.

In fact, it is almost certain that Bill never had the diagnosis of “Alzheimer’ disease” (AD) to begin with – this is because doctors do not usually make this diagnosis. The diagnosis of “Alzheimer’s disease” is based only upon a brain biopsy (or autopsy after death), but brain biopsies are rarely done to diagnose dementia. Instead, we diagnose “Alzheimer’s type dementia,” which means that someone has dementia of a type compatible with AD and other causes have been ruled out – it is a diagnosis of exclusion, and a clinical diagnosis. But that diagnosis is only as good as the thoroughness of the workup to rule out other causes. In addition there are disorders that can mimic AD unless a careful assessment is made.  Even in the best circumstances, there will be the occasional misdiagnosis simply due to the limitations of our ability to diagnose AD.

So the first question we have to ask is – did Bill actually have AD or something else that was reversible? We cannot know, and we are not given sufficient information to make this judgment. It is even possible that Bill had a nutritional dementia – there are some vitamin deficiencies that produce dementia that can mimic AD, like B12 deficiency. So there will be some patients who improve with supplementation. This is not a “miraculous nutritional cure” for AD, but rather just treating a nutritional deficiency that causes dementia (but not AD).

Another common disorder that mimics AD is what is called “pseudodementia” – this means that the brain is not functioning well, but there is no biological damage. This can be cause by sleep deprivation, or depression. Pseudodementia is 100% reversible because there is no permanent damage to the brain.

So it is possible, even probably, that Bill never had AD and his anecdote is therefore not relevant to AD.

But I also don’t buy the claim that he is improved. He may be, but the videos are not impressive. You may notice that Bill says very little during these videos – Nita does almost all of the talking. He does relate some historical tales – exactly the kind of long term memory that is selectively preserved in many dementias – but does not give evidence that he does not have some impaired cognition. It may just be their personalities and relationship that cause Nita to do almost all the talking – but this is also a very typical dynamic for a couple where one member has impaired cognition. I am not making any judgments about what Bill has or does not have from this evidence, just pointing out that if the point of the videos was to convince us that Bill is better, they failed utterly.

Finally we are told that a very low carb diet and coconut oil can cure AD, but doctors are just not interested because it does not make them money. This is naive and insulting, of course. First – treating dementia is not a money-maker for doctors. It is just about the least lucrative disorder that neurologists treat. From a purely selfish point of view, we would love to be able to cure AD with a simple dietary change. And, I think that most doctors are not heartlessly selfish and would actually like to help their patients with dementia. In fact we routinely diagnosis and treat B12 deficiency, so if AD were a deficiency of saturated fats why wouldn’t we treat it the same way?

In the end, simplistic conspiracy theories are not compelling – the scientific evidence will determine how AD is treated. If anyone can provide compelling (i.e. not anecdotal) evidence that coconut oil helps AD then we will all prescribe it. So far there is no such evidence – only self-promoters selling anecdotes and conspiracies. A PubMed search on coconut oil + dementia or Alzheimer’s disease yielded exactly zero results. Google searches yield only promoters – no evidence. I also find it very telling the the promoters don’t link to any actual published evidence – they probably would if it existed.

24 responses so far

24 thoughts on “Alzheimer’s and Diet”

  1. Draal says:

    Dr. Novella,
    If this story didn’t include “coconut oil” but rather “fish oil”, would your assessment change?
    ISI Web of Knowledge (all databases):
    coconut oil dementia = 0 results
    fish oil demetia = 56 results
    Draal, Ph.D.

  2. B Hitt says:

    Whenever the word “cure” is tossed about cavalierly (particularly in the context of a disease such as AD), it’s a sure sign of pseudoscience. Complex, chronic illnesses are usually treated, managed, or controlled, and the word “cure” has a special meaning and is used in limited types of situations. Certainly there may be a cure for many of these complex diseases one day, but (at least for a disease as poorly understood as AD) today is not that day. I feel like I’ve sat through entire days of talks at AD conferences without hearing the word “cure” used once.

    I wonder how many pushers of pseudoscience haven’t caught on to this and don’t realize how much this undermines their ability to sound legit and scientific, and how many see this as an opportunity to get a leg up on science-based medicine with their rhetoric. They must catch on a little. There was the infuriating time that an accupressure/herbalist acquaintance told me that in the CAM community they know that sage prevents Alzheimer’s, magnanimously pointing out that they say “prevent,” not “cure.”

  3. One huge red flag here is the idea that a treatment for Alzheimer’s wouldn’t be lucrative to the doctors who came up with it. If you want a license to print money, cure something that affects mainly the elderly.

  4. B Hitt says:

    Salient point from the fish/coconut oil comparison:

    -Fish oil is pulled out of the sea and tested for its effect on dementia (with mixed results).

    -Coconut oil is pulled out of someone’s ass and enthusiastically celebrated as a miraculous cure for dementia without rigourous testing.

  5. SpicyCupcake says:

    @Draal I can field that question Dr. Draal. Yes, it would change to a review of the evidence instead of a frustration with an utter lack of evidence to support the claims being made. I don’t speak for Dr. Novella but he has an established history of not being a priori about subjects. I think this post was an honest attempt to review what evidence was available. Unfortunately it was really only anecdotal.

  6. Watcher says:

    Lets not forget that some dementias can be “self-limiting” so to speak. Cognitive impairment might begin, but never progress past a certain stage into a full blown Alzheimer’s-like disorder.

  7. Draal says:

    “Yes, it would change to a review of the evidence instead of a frustration with an utter lack of evidence to support the claims being made.”

    I suppose I should admit I did a little digging a few minutes ago and answered my own question, in part. Coconut oil does not contain the fatty acids (like DHA) which are found in fish oil and are the focus of several recent studies. In other words, not all oils are created equally.

    Still, I’d be interested in a skeptical review by Dr. Novella on the fish oil studies for dementia since Steven is a neuroscientist and could break-down the literature much more effectively than I could.

  8. ccrome says:

    My son got a quite rapid worsening in tics (pretty severe Tourette type tics), and man, nearly everybody asks, “can you help that with diet?” I’m sure that there are lots of things that can be helped with improved diets, and perhaps TS is one of them. Of course, my son is 10 (in the typical range for TS severity increasing).

    However, the ‘everything-can-be-cured-by-diet’ reaction seems to be pervading the collective psyche. It’s become a sort of cure-all for anything that people don’t know how to cure.

  9. HHC says:

    Why would adding a saturated fat help demetia? The elderly often have high cholesterol. Won’t a diet high in saturated fat boomerang? You would want a licensed dietician to prescribe a proper diet to meet the individual’s needs.

    Vacular dementia is an alternate diagnosis which may describe the Alzheimer’s type disorder. Psychiatric intervention to treat any underlying emotional disturbance is important. Problems that have survived with the patient need to be addressed especially when they leave their dysfunctional family and enter 24 hour care facilities.

    How do you change an Alzheimer’s diagnosis? I simply requested that the change be made based on the science information from Dr. Novella’s blog. Thank you.

  10. The science of the ketogenic diet and ketones is an interesting topic I did not have time to explore with this post.

    There are documented neuroprotective effects of a ketogenic diet, but it is not clear which metabolic elements are important. Also it’s not clear if the effects are symptomatic or disease modifying – meaning that ketones may simply provide energy to neurons starved by impaired glucose metabolism.

    This is all interesting but preliminary, and needs more study from basic science to net clinical effects.

    This also has nothing to do with coconut oil, which by itself would not produce a ketogenic effect.

  11. SARA says:

    Apparently if you want to sell a book, pill, powder, or snake oil – you merely have to pick a random exotic sounding food and make up a good story. Although in her defense, I think she thinks its true.

    I would actually be less sad if she was doing this in full knowledge of scamming people than if she has convinced herself that this entire thing is true.

  12. tmac57 says:

    I’m afraid that I will have to disagree with Dr. Novella on this point:”In fact, it is almost certain that Bill never had the diagnosis of “Alzheimer’ disease” (AD) to begin with – this is because doctors do not usually make this diagnosis. The diagnosis of “Alzheimer’s disease” is based only upon a brain biopsy (or autopsy after death), but brain biopsies are rarely done to diagnose dementia.”
    My mother suffered from vascular dementia (she had numerous TIAs during a 12 year period before her death),but that didn’t keep about every other physician that she encountered(and there were very many)from referring to her “Alzheimer’s disease” even after I would correct them by saying that she had been diagnosed with vascular dementia.After she died,I obtained the death certificates,and on the cause of death they had put ‘end stage Alzheimer’s’ (no autopsy was done).
    Had I not known better, I would probably have picked up the Alzheimer’s diagnosis from the fact that they kept calling it that,and whenever I had to fill in a medical history for her I would have repeated the error.

  13. Collden says:

    Steven, the compound used in the study, AC-1202, is the medium chain triglyceride caprylic acid. Coconut oil is 8% caprylic acid by weight, as well as 6% capric acid, which is almost as ketogenic. Add in the 45% lauric acid, which is also somewhat ketogenic, and I don’t think its implausible that you could achieve the same ketogenic effect with a reasonable amount of coconut oil, as they did in this study with their 20g of capric acid.

    Anyway, you weren’t only talking about coconut oil, but also about low-carb diets, and a low carb diet could certainly be highly ketogenic by itself.

  14. daedalus2u says:

    A ketogenic diet does increase NO levels.

    (note the title of this paper is actually the opposite of what they found. They found increased NO and they found increased time to KA induce seizure, that is the increased NO prolonged the time until the seizure happened).

    This is probably the mechanism behind the improvement in AD. A ketogenic diet can be quite difficult to maintain. In the treatment of epilepsy the state of ketosis must be quite strict, relapsing to a carbohydrate containing diet can cause relapses of seizures very quickly (less than an hour).

    In this trial, there were more withdrawals due to side effects in the treatment leg.

  15. sonic says:

    I’m wondering who and how it was determined the guy had AD.
    I hear people say “so and so has AD,” regularly based on the fact some older person forgot something.
    The word has a cultural meaning that is quite different from the medical one.

  16. The relationship between consuming coconut oil and ketogenesis is complex and unclear – so that you cannot extrapolate from this study to the addition of coconut oil to the diet.

    What we are left with are no studies of coconut oil directly, and wild clinical speculation from preliminary data.

  17. Woody says:

    The article Collden posted above is the basis for the FDA approval of Axona as a medical food for Alzheimer’s disease (AD), and is an example of researchers bypassing rigorous peer review in order to make a profit. It is much easier to get something approved as a medical food than a medication. The medium chain triglycerides in that study are derived from coconut oil, at least in part.

    There is a wealth of data supporting the idea that altered glucose metabolism in the brain is a major factor in the expression of AD symptoms, so it is reasonable to look at alternate fuel sources for the AD brain such as ketone bodies. The problem is that the Axona folks ran with this single study without doing any replication or more definitive research. Another key finding in this study that is glossed over in the Axona marketing is that APOE4 allele carriers did not benefit, so even though they market the compound to all patients suffering from AD, probably only half of the AD patients would stand to benefit based on the available scant data.

    I would also disagree somewhat with Dr. Novella’s characterization of how AD is diagnosed. The appropriate terminology is “clinically probable AD” or “clinically possible AD” when other dementia mimics have been excluded. The other issue is that AD outstrips other causes of dementia by a large margin when one looks at clinicopathologic series of dementia. The second and third most common dementias, dementia with Lewy bodies and vascular dementia, are often found to have comorbid AD pathology at autopsy. So basically, if someone were to diagnose dementia, they could blindly label the underlying cause as AD and be correct the majority of the time.

  18. SimonW says:

    Coconut oil gets a lot of pushing in the world of thyroid disorders. Usually claims relating to weight loss.

    In this area, the task of finding relevant research papers is particularly challenging, as coconut fats are often used to fatten up rats for research into weight loss. One would think that this fact alone would discourage the promoters, but no they always have a reason why we aren’t like lab rats.

  19. Calli Arcale says:

    SimonW — except, of course, when they want to use studies in rats as evidence *for* various interventions. (e.g. “Rats fed this had fewer dementia symptoms!”)

  20. Collden says:

    There is a dearth of studies on the effects of coconut oil in humans in general. For some reason all the research is focusing on purified medium chain triglycerides, which is pretty consistent in showing beneficial effects of MCTs on weight loss, energy expenditure, blood lipids, etc.

    I also question the idea that coconut fat stands out as a particularly fattening substance in rat models. Corn oil is far more popular for the purpose of “high-fat diet induced obesity”. In fact, coconut oil has been found to be a poor choice in rat models of diet-induced obesity, since it does not seem to cause the usual negative metabolic effects typical of rats fed high-fat diets.

  21. tmac57 says:

    Since this thread has turned to coconuts, has there been any epidemiological studies on populations that consume them in regards to AD?

  22. HHC says:

    tmac57, To answer your question, studies could be done on human populations where coconut production is the greatest, which is the Phillipines, Indonesia, and India. But I don’t think physicians in these areas believe that AD is a pressing issue. In India and Indonesia, open defecation is the basic health problem.

    Coconuts are produced in the U.S. states of Hawaii, southern Florida, and the U.S. territories of Puerto Rico and the Virgin Islands. Did you know that 100g of raw coconut meat contains 29.7g of saturated fat of the total 33.49g fat?

  23. nwmslaw says:

    Um, I really skeptical of your claim that a ketogenic diet, whether using MCTs such as coconut oil, or a modified Atkins diet, has “no” support for the treatment and/or prevention of Alzheimer’s. Of course, longitudinal studies that definitively prove the claim that ketogenic diets are useful in AD in humans are by definition elusive–they’re longitudinal studies after all. But for starters–and this was from a .06 second google search, see: and And check out Dr. Emily Dean’s blog at, where you’ll find some pretty compelling arguments (but no longitudinal studies, of course) for the use of such diets. Is it not true that glycogen metabolism is compromised in AD patients, and that ketone bodies are an alternative source of fuel for most areas of the brain? In any event, my extensive anecdotal experience with most MD’s is that they know painfully little about nutrition, and are only interested in pharmaceutical interventions. That’s the fault of the med school curriculum. The best approach would be to at least try such a diet in clinical practice with AD patients, and see how it goes. What’s there to lose, doc?

Leave a Reply