* Behavioral studies of mice can be very subjective in their own right, so you can definitely get bias, particularly with small N.
* There's no generally accepted model of autism in mice. In other words, mice might not even be capable of getting autism in the first place, so saying that "these behaviors in mice correlate with autism in humans" is a bit of a limb.
* No measurement of the microbiome in the descendants-of-the-implanted-microbiome (the ones that were actually tested) was demonstrated.
So you can come away from the paper with the feeling that they set out to test a hypothesis, interpreted subjective results in the way that best matches the hypothesis, and failed to (or perhaps even purposefully excluded) test for alternate hypotheses that might explain the evidence better.
Why do you focus on the mice, when the article talks about two studies, one of which was longer term and done on humans?
"Crucially, these changes in gut bacteria have translated into behavioural changes. Even 18 weeks after treatment started the children had begun showing reduced symptoms of autism. After two years, only three of them still rated as severe, while eight fell below the diagnostic cut-off point for asd altogether. These eight thus now count as neurotypical."
Because he knows something relevant about the mouse study, but not the human study. He's not objecting to the conclusions or tone of the Economist article, just the mouse study. Seems legit to me.
I was half-amused, half-horrified at the remote possibility that vaccines might alter the gut microbiota and actually cause autism. I couldn't bear hearing "I told you so" from my antivaxxer friend.
_sigh_ this feeling again when you read about some great "paradigm-shift" advancement and then go to HN comments and find out the study has bunch of holes. :(
No the research is getting pretty definitive in this area, don't count it out and the article also specifically talks about another study not the mouse one where they had real world success (over 50%) treating autism in people via bacteria transplant.
the very small N part still applies here though, the study was done on 18 children only, given the prevalence of ASD and the complexity of it I think we just need more proof. I won't count it out but I think the conclusions should've just been: this looks promising but we need to study it more. (tbf, i think i'm complaining more about the tone of the article.)
The sad thing is that in America today you can't get access to the two big 'miracle cures' (fecal transplants or MDMA) for any reason outside of clinical trials. Anyone with a non-verbal loved one should be able to legally try the fecal transplant and anyone with PTSD should be allowed to legally try MDMA. Scientifically proving that treatments work isn't nearly as important as proving that treatments are safe.
Yep, it's totally likely they became neurotypical all of a sudden for some other random reason. If several small studies are all pointing in the samd direction, something is probably going on. Meta analysis dude.
There were no controls, and autistics frequently get better at managing symptoms to point of making the diagnosis harder. Why isn't there a high quality study of this?
I think something which should be considered, is that gut bacterial abnormality is very plausibly a consequence of autism, rather than a cause. Selective eating and food refusal occurs in about three quarters of children with ASD, and has been used as a diagnostic criteria (which reinforces the correlation). Poor diversity in nutrition leads to poor biodiversity in the gut flora.
That said, poor gut biodiversity could in turn lead to exacerbated presentation of symptoms, due to gut flora's involvement in neurotransmitter systems, interactions with the gut's nervous system, or simply painful intestinal distress.
I wonder if there should be a similar account for "in males".
I believe many medicines are tested in males only because it would be unethical to test them in possibly pregnant females. However, one ridiculous example is birth control pills.
> Crucially, these changes in gut bacteria have translated into behavioural changes. Even 18 weeks after treatment started the children had begun showing reduced symptoms of autism. After two years, only three of them still rated as severe, while eight fell below the diagnostic cut-off point for asd altogether. These eight thus now count as neurotypical.
If it works both directions - almost makes me wonder if there are bacteria that can manipulate our brains (causing picky eating) to get their preferred food and keep themselves in the majority in the gut.
Regulatory capture for bacteria.
I mean, look at what Toxoplasma gondii does to mice...
Fortunately, they're not influencing gut flora research to make humans treat some species preferentially. This would be way too complex for simple creatures that don't have their own nervous system.
Another straightforward possible mechanism is that upset stomach exacerbates behavioural systems.
Consider anything you do that takes conscious effort, in life or work. Now try doing it with a headache, or on lack of sleep, or with a stomach pain. You'll do worse!
I mean, yes, but to go from severe ASD behaviors to neurotypical is vastly different than perhaps some behavioral artifacts from a headache or stomach pain. I think your conclusion is greatly understating the result.
I wonder whether much of the difference in severity of autism symptoms, rather than being differences in the underlying condition are explained by differences in comfort. Symptoms used to differentiate sever cases such as irritability, inattention, repetitive behaviour, mood changes, sleep disruption, and hyporeactivity are all exhibited by neurotypical people when in pain / discomfort / stress.
A core feature of autism is sensory abnormality, so perhaps the symptoms present in severe cases can be explained by increased sensitivity to these negative internal stimuli.
Additionally, I think it's important to note that we would discount many of these symptoms from a diagnosis, if a more observable and salient cause (e.g. an itchy skin condition) were seen. Gastrointestinal discomfort being poorly observable creates a bias.
> Selective eating and food refusal occurs in about three quarters of children with ASD
I’m sure there are differences between ASD and non-ASD kids in terms of diet, but picky and selective eating is present in a lot of children that are never diagnosed as ASD.
I am not a microbiome expert but I have always wondered how dynamic is the gut microbiome? For example, maybe I travel for a week and consume a bunch of fast food and drink a bunch of alcohol or I contracted food poisoning. Would my gut microbiome be substantially different than if I followed my regular diet of relatively healthy home made foods? and how long would it take to go back to its default state before I travelled (if ever).
I've been told (by someone who has had C. diff and dealt with a number of gastro issues) that it can take up to 7 years for your gut to stabilize after simply taking antibiotics.
With that being said, some googling just now to make sure I'm not just spouting nonsense hasn't turned up a 7 year figure. But it has returned a bunch of articles pointing to studies that suggest it takes a long time. [1] says after 10 months there's still a disturbance.
I have seen fairly convincing evidence presented quite recently that for the vast majority of people it may take months but after that period the intestinal microbiome is nearly indistinguishable from how it was before administering antibiotics.
There was an interesting post on Reddit recently by someone who tracked their gut microbiome composition over a three-year period, with some nice graphs [1] and source code.
The data were compiled with uBiome SmartGut and Explorer
kits over a 3 year period. I directly parsed the .json
files they provided. The plots were generated with
matplotlib and the source code is available here:
https://github.com/isaacgerg/ubiome_longitudinal_analysis
Not sure how scientific this is, but it was interesting to me that someone would even do this.
Different bacteria thrive on different nutrients, so changing your nutrient intake will affect their populations. I wouldn't expect a week of fast food to be significant compared to going on a ketogenic diet (starving all carbohydrate eating bacteria). Also, intake of new bacteria affects them, as it changes the competitive landscape.
Taking antibiotics massively reduces the quantity of gut bacteria. It often radically changes the balance as their numbers recover, resulting in a proportional increase in resistant strains.
Wonder what is the impact for those who just take Soylent for their meals. Does the lack of variety of nutrients kill of some bacteria in the gut and are they worse off for it?
I'm not sure it's correct to say Soylent constitutes a lack of variety of nutrients. Soylent and similar products contain a balance of macronutrients (carbohydrate, lipids, amino acids), vitamins and minerals.
Eating a ketogenic (almost no carbs) or fruitarian (almost entirely carbs) diet, is much more likely to create an imbalance in types of gut flora.
Yes. There's a long list of different starches and other molecules (e.g. connective tissue in joints) that your gut bacteria feed off of, and different strains of bacteria prefer different types of food. In the FODMAP diet, you avoid certain starches in a deliberate attempt to shift your biome diversity.
I hope this doesn't provide ammunition for the people pushing making autistic children drink bleach or giving them bleach enemas as a "cure" for autism. Yes, there are such people, and yes, there are parents stupid enough to take their advice and do that to their kids.
Since killing bacteria is one of the legitimate functions of bleach, I can see them claiming that their bleach therapy will work by killing off that bad, autism-linked gut bacteria.
If it's true, it will be one of the biggest medical breakthroughs!
Shifts like this happen and lead to huge difference in people's lives. Now mostly forgotten, but it was believed that ulcers are caused by stress and food and not bacteria. I have a family member that didn't survive surgery for mistreated ulcer. Then, for me it was a simple matter of taking antibiotics for few weeks.
The most robust way to deliver new bacteria to your gut is with a gastro-resistant capsule (so it bypasses your stomach acid). Food will introduce it too; fermented foods contain more.
Introducing the bacteria is only part of the problem though. If you want to have a diverse community of gut flora, you need to have a nutritionally balanced diet.
Fecal transplant. There's not really any other way and most "advice" you're getting is regurgitated unscientific, data-free hogwash from the colon cleanse and health fad industries.
Second that for sauerkraut and basically any pickled food.
For probiotic foods to have any meaningful effect, though, make sure your gut is healthy first. Probiotics foods are good for one type of bacteria but you also need to neutralise the bad types. Stop eating sugars, saturated fats etc. Your stomach is an ecosystem that will suffer any changes quite badly at first, but will thank you in the long term.
I chose to be vegetarian aged six, this was probably my first conscious thought as opposed to doing what I thought other people wanted me to think. I am sure I could put meat in my mouth if under duress but I am fairly sure that the gut bacteria most people have for digested meat have died out a long time ago, I have effectively starved them.
I am no scientist and I am not wanting to push a diet choice on anyone, I went vegetarian simply because the food was better at school for the few that were specially catered for as vegetarian, with larger portion sizes being key to my motivation at the time.
However, as a science experiment you could go vegetarian for a year to selectively cull the bacteria that go with eating meat. You could go full on vegan too but that isn't as easy a change as going merely vegetarian.
Expect some flatulence that will sort itself out in time, something you might also want to look into as part of your experiment.
Allegedly people in Africa who actually eat have the most diverse gut bacteria and these guys ain't eating hamburgers.
I think that a change over to full vegan would be with health negatives to consider, a change to vegetarian would not run that risk. It is worth remembering that anyone who is anyone in Hollywood is in the serious business of looking after their looks, they tend to be not eating steaks. In sport there are also athletes that only hit the high notes of performance after going on a mostly plant based diet.
The probiotics business is a business, they have something to sell. Nobody except for book authors are going to sell you the idea of removing rather than adding something from your diet.
As mentioned I am vegetarian by chance - a school dinner lady just happened to do much nicer vegetarian food than meat eater food and I blatantly lied on my first day at that school about being vegetarian just to benefit from the better portion sizes. I had to stick to that lie thereafter and my parents were okay with me not having meat at home. I don't have any skin in the ethics game unlike some recent converts to veganism. However, that said, I can see the upsides financially whenever I dine out. Although bills get split I am aware that my part of the bill is 2/3 the price of whatever everyone else is having or even less than that. I am okay with a bowl of pasta whereas others around the table have some endangered species from some ocean on the far side of the planet, it is obvious what is going to be the cheaper option.
You can conservatively estimate four figure savings by spending a year in vegetarian mode so maybe think of something you want to put that money towards in order to cope better with the hardship of not scoffing burgers etc. for a year.
This whole comment seems to be based on the premise that eating meat leads to bad gut bacteria. Do you have any sources for that? I am genuinly curious. My mental model (which isn't based on much hard evidence) is that sugar and processed foods are the things to avoid when aiming for good gut bacteria.
Nowhere am I saying that meat leads to bad gut bacteria, what I am saying is that you can starve out the gut bacteria that eat meat and live for a while with just the bacteria that eat plant based foods.
Because you can't live off just meat alone (unless you are a dog or other carnivore with a short intestine) there isn't the option to cut out the plants and just live off burgers and sausages for a year, to clear out and reset the gut bacteria that like a bit of plant material. Going vegetarian is therefore a safe personal experiment.
The problem with processed foods - even if natural and without additives - is that they don't make it past the small intestine with any nutrients in them for the gut flora of the lower intestine.
Vilhjalmur Stefansson demonstrated this in the USA a century ago, but this was based on his experiences watching and living in the arctic where we have much longer natural experiments to refer to.
Of course, whether you should or not, and the nature of the long term effects, are separate considerations.
I responded to a comment about how to get good gut bacteria which was asked in response to the original article. I contributed an original idea that by doing A/B testing on one's own gut you can starve half the bad bacteria.
I qualified my idea by stating that I don't eat the dead stuff.
It is cool you can write snarky and pedantic comments but it has nothing to do with sharing ideas which is what other people like to read.
I probably should have bit my tongue, but it seems like one can't read a single comment thread these days without tangents about veganism or the wonders of a plant-based diet turning up like dandelions after a summer rain. Evangelism gets my back up.
The biggest issue with a lot of autism research is in how they measure the symptoms of autism. Autistics will often "melt down" or engage in other stereotypical behaviors when the environmental stressors are more severe than what they can deal with through behavior that appears "normal". The more severe the stressors, the more severe and frequent the behavioral deviations.
In short, most scientific approaches will often conflate reducing autism with helping people handle their autism better. It's a pretty chronic problem for understanding the disease, but much less impactful for generating practical treatments. Mental healthcare in general doesn't care much whether it reduces the severity of neurodivergence or the distress and socially problematic behavior caused by neurodivergence.
This might sound naive, but isn't the easiest method to reject the hypothesis is to gather data on gut bacteria in people with ASD and those that don't? Wouldn't that be much more direct and explain much more? Or is there some sort of technological or cost limitation to this method?
Call me old-school, but from my perspective, if we are talking about actual evidence, you need exactly one piece of actual evidence, and a case is settled.
No, we are at peak “gut bacteria” science denial. On the other hand study of the microbiome is still in its infancy and looking very promising vis-a-vis human health.
Replying to you since this is the top response but aimed at y'all really.
At the risk of stating the obvious. Gut bacteria are bacteria... in the gut. Like if you tell me they affect the absorption of nutrients - I can see that. Maybe they make something more bioavailable or maybe they break something else down. If you tell me their involved in deficiencies and over/underweight that kinda makes sense.
Tell me they affect appetite through manipulating our signals ok. Not as obvious but not implausible either.
Tell me they cause highly complex behavioral differences... it just seems implausible. But maybe gaba can indeed explain it.
People used to say the same thing about washing hands -- "little organisms on my hand are making me sick? I don't buy it."
Nowadays, the gut is sometimes referred to as the "second brain" because of your enteric nervous system which is located inside the gut. It probably has more influence over you than you're aware of, as it operates independently of your CNS.
>> People used to say the same thing about washing hands -- "little organisms on my hand are making me sick? I don't buy it."
If you're talking about the time before there was strong evidence for pathogenic bacteria- then that would have been the only rational and justified behaviour.
Why should anyone believe something that sounds implausible given current knowledge, before strong evidence is presented to support it?
I'm talking about when evidence was first being discovered, as is the case in this thread. Humans have a disconfirmation bias - we have a hard time accepting new, conflicting information.
Yes, well, why should we accept new, conflicting information? The reasonable thing to do is to wait before you have a good enough reason to accept a new thing.
The study of the interaction between microbes and the human body is not 3-d printing or SDCs. Scientific fields of research feature their share of hype, but they don't run through product cycles like tech companies deal with. This is due to a number of factors including different standards for objectivity, transparency, financial interests, and at the heart of it: we are much more likely to devoutly wish for medical breakthroughs in cancer or Alzheimer's, whereas there are relatively very few people praying for Google Glass or VR, save those trying to induce demand for such products.
So, the hype cycle for tech products is not really as useful in the study of microbes in the human body.
I heard that gut bacteria determined which type of entertainment you enjoy. HBO has been secretly subsidising fast-food chains for years in order to keep the price of burgers down and so get more people to watch violent shows like GOT.
It is true there have been hype waves. But there also have been huge medicinal revolutions too (e.g. vaccines, antibiotics, etc)
It sounds to me that you're just lumping all "great claims" into the "false" category with no basis. It doesn't sound like you read the studies and take issue.
There appears to be an increasing amount of evidence linking an unhealthy gut microbiome with various conditions (mostly psychiatric), including a few big ones like Alzheimer's and depression. This has gotten me wondering: how many people might be sick as a consequence of their poor diet? It's probably a good idea to carefully scrutinize your diet as a potential influencing factor if you have a medical condition. Maybe making a few tweaks could end up helping you feel better? In any case, always discuss things with your doctor!
Anecdotally, I began to regularly consume more fermented foods (e.g. plain Greek yogurt, kimchi, kombucha, kefir) about 8 or 9 months ago, which helped trigger a cascade of other positive changes and led to gradual but marked improvements to my health.
> This has gotten me wondering: how many people might be sick as a consequence of their poor diet? It's probably a good idea to carefully scrutinize your diet as a potential influencing factor if you have a medical condition.
If we separate microbiome from diet a little here, then it's quite easy to say that many people are sick because of their poor diet. Heart disease and cancer are by a large margin the largest killers of people in the United States, both of which have obesity as a major factor. Type 2 Diabetes, high blood pressure, and a host of other diseases also result from poor diets. I'm practically certain it's one of the primary causes of premature death.
> This seems to be a trend: a study finds correlation between X and Y in mice, no evidence that X is linked to Y in humans.
There are lots of regrettable trends.
The story posted is primarily about the results of experiments with 18 humans between the ages 7-16, over two years, and a section at the end describing some earlier studies involving mice.
The results of the human experiments are astonishingly positive.
I mention this just in case you breezed too rapidly over those large parts of the article.
The salient points are:
* Very small N.
* Behavioral studies of mice can be very subjective in their own right, so you can definitely get bias, particularly with small N.
* There's no generally accepted model of autism in mice. In other words, mice might not even be capable of getting autism in the first place, so saying that "these behaviors in mice correlate with autism in humans" is a bit of a limb.
* No measurement of the microbiome in the descendants-of-the-implanted-microbiome (the ones that were actually tested) was demonstrated.
So you can come away from the paper with the feeling that they set out to test a hypothesis, interpreted subjective results in the way that best matches the hypothesis, and failed to (or perhaps even purposefully excluded) test for alternate hypotheses that might explain the evidence better.