Hacker News new | past | comments | ask | show | jobs | submit login
Fermented foods, neuroticism, and social anxiety: An interaction model (nih.gov)
105 points by nyodeneD on June 11, 2015 | hide | past | favorite | 48 comments



If you are curious about the foods:

> To determine participants׳ consumption of foods, they were first asked to think of their food intake over the past 30 days (see Appendix). This was followed by a list of 10 items consisting of the following: 1.fruits and vegetables of all kinds, including fresh, canned, frozen, cooked, raw, and juices; 2. yogurt, 3. kefir, or food or beverages that contain yogurt; 4. soy milk, or foods or beverages that contain soy milk; 4. miso soup; 5. sauerkraut; 6. dark chocolate; 7. juices that contain microalgae; 8. pickles; 9. tempeh; and 10. kimchi. Participants were asked to indicate how often they consume each of the foods using the following 7-point scale: 1 (never); 2 (1–3 times in the past month); 3 (1–3 times per week); 4 (1–3 times per day); 5 (3–5 times per day); 6 (5–7 times per day); and 7 (more than 8 times per day). Scores were converted to monthly frequencies of 0, 2, 8, 60, 120, 180, and 240, respectively.

And the Conclusion:

> This study provides the first connection between natural fermented food consumption patterns and anxiety. Increased consumption of fermented foods likely to contain probiotics was associated with fewer social anxiety symptoms. This effect was qualified by an interaction with neuroticism; those who were highly neurotic showed fewer symptoms of social anxiety with greater consumption of fermented foods. Also of note, increased frequency of exercise was associated with decreased social anxiety. Taken together with the results of previous preclinical and clinical studies (Rao et al., 2009, Silk et al., 2009, Messaoudi et al., 2011 and Tillisch et al., 2013), the current results suggest that fermented foods likely to contain probiotics may have a protective effect against social anxiety symptoms for those at higher genetic risk, as indexed by trait neuroticism. While a follow-up prospective cohort study or clinical trial is necessary to determine the direction of causality, these results suggest a possible clinical benefit of fermented foods, a low-risk intervention, on social anxiety.


Everything about this study says it is not worth taking seriously. Why? Because...

1) The size of the effect is not reported. Without any indication of what the dose-response curve is other than it has a positive slope (at what significance level?) we know nothing about the relevance of this result to clinical practice.

For example, it is true that if you pay very careful attention to the sidewalk you will increase the amount of dropped money you find. It's not going to make anyone rich, though, any more than just any measurable effect is going to make one less anxious to a useful degree. There is zero indication of how big the effect is, other than it is measureable in a large population. The lack of indication of effect size suggests that it is really small.

2) Self-report via recollection is a lousy data collection technique. Was a food log app really too much trouble? So the raw data are incredibly noisy and may contain systematic errors that are related to the phenomenon under test.

3) They've subsetted the data by pulling out neuroticism, thereby dramatically reducing the significance of any result by an unknown amount (how many other characteristics did they investigate?)

This is the greatest sin. Any time you see a paper that says, "The effect was particularly large in sub-population X" the first question to ask it, "How many sub-populations did you look at?" The next question to ask is, "What is your calculated probability that any one of your sub-populations would have had such a large effect by chance?" These numbers should be reported in the abstract because people subset their data all the time and then analyze the subset that shows an effect as if it was the only one. This is either incompetence or dishonesty.

Statisical significance rolls off with astonishing speed as you subset the data, and if you don't have a formal discipline for subsetting as part of the experiment design that is fixed prior to analysis, any result on a subset is necessarily insignificant because you could have kept subsetting until you found a 'significant' result. If you don't have a formal discipline that says, "We will cut the data in these N ways and only these N ways" you are going fishing on an infinite ocean, where you will certainly catch the kind of fish you are looking for.


Oh yes, I agree.

More importantly, food is inherently cultural and varies on class and it might be that neuroticism/social anxiety does as well. I posted this because I thought it sounded like a week study with an overly strong conclusion.


>I posted this because I thought it sounded like a [weak] study

Let me see if I understand: you posted (the abstract of) a study you consider not worth taking seriously so that you could read criticisms of the study by people who know more than you; is that correct?

If not, then please explain more about your motivation for posting because I'm a little confused on that point.


batbomb did not post the paper.


Given the effect probiotics have on one's gut flora, and the newly-discovered pathway from the gut to the brain via the lymphatic system, this seems to be in the realm of possibility.

http://neurosciencenews.com/lymphatic-system-brain-neurobiol...


In rats, at least, the vagus nerve is a link.

Supplementing a rat's diet with Lactobacilus Rhamnosus reduces their anxiety level.

Cut the vagus nerve, the effect disappears.

http://www.pnas.org/content/108/38/16050

This doesn't mean that other bacteria may affect the brain through the immune system, though.


There's already a huge link between the gut and the brain called the vascular system.


Subject to the brain blood barrier, yes.


I'm curious, how is the lymphatic system a connection between the gut and brain?



It's recent news that the lymphatic system include the brain, but I still don't see the connection with the gut.

Generally lymphatic systems flow away from organs (the lymph vessels in your arms flow back to the heart). How would something in your gut travel to the brain through the lymph system?


This is a correlational, not causative study. It's possible there's some third factor that cause both an increase in eating fermented foods and a decrease in social anxiety. Amount of free time, for example, or belief in healthy eating, or vegetarianism, or income. They do try to control for confounding variables (fruit & vegetable consumption) but there's no way to control for everything.

Let's wait until someone conducts a double blind study before drawing too many conclusions from this.


One nice thing about this claim: it should be easily testable. In fact, someone could test it themselves; score yourself on a Big Five inventory, if you're >75% N, make some capsules with probiotics (I believe they come in powders so this is easy), take them blind, go to a party or other social interaction, and self-rate how stressed you are.


Do you expect it to be that quick acting? The study seemed to be what they ate the last 30 days, so I imagined an effect that slowly built up. At the very last some time should be allowed for the bacteria to reach the gut and start colonizing it I think? Also with regards to parties, my experience is that alcohol can affect the gut flora.


The dose can be taken in the morning, given it much of the day to act. Acute effects seem plausible here given the neural connection to the gut, however, a self-experiment examining acute effects only is still useful to help rule that out. And to investigate slower effects, you can block the doses so instead you take a week (or a month) at a time (this is more work because you need more blocks to achieve similar numbers of units as a per-day randomization).

> my experience is that alcohol can affect the gut flora.

So? You don't know if the alcohol effect would be to enhance or reduce the supposed calming effect or not affect that at all.


Instead of capsules can I make kimchi and eat it for a week? Something that calls for fish sauce and hot peppers is bound to have some social impact.


Sure; just note you won't be proving fermented foods lessen social anxiety -- at most you can conclude that consuming kimchi does it.

It's pretty good so you're not at loss anyways! I find sauerkraut enjoyable too, vegetables are too bland for me, a little bit of this stuff is a sauce replacement for me.


Unless you know a vegetarian, you may have trouble finding gel capsules sufficient for a decent dose of kimchi... (Also, now you have to refrigerate all your capsules.) Easier to just use some probiotic powder stable at room-temperature unless there's some compelling reason to expect them to be ineffective.


Vegetarian?


Er, brain fart. I meant 'veterinarian', so you can get horse pills...


Only if you can find some non-fermented placebo kimchi that looks and tastes just the same, otherwise it ruins the blindness of the experiment. Hmm, Placebo Kimchi sounds like a good name for a band :)


There's one guy who has experimented with 'megadoses' of probiotics for curing social anxiety: http://www.scribd.com/doc/28819342/Experimental-Treatment-fo...


I don't see how one person's blinded self-experimentation would yield more knowledge in this case, as the proposed experiment is not methodologically sound and as the data would not tell you anything more about the effect - as far as I can tell, anyway.


Except that's unblinded so placebo would probably dominate over actual effects. You'd have to make a placebo pill as well and not know which one you were taking until after you gather all your results.


That is what I meant, yes. Please forgive me for not including a full description of my self-blinding method in a quick comment.

> until after you gather all your results.

Not all of them; you only need to be blinded for each unit. Once the data is collected for that unit, then you can unblind yourself (and this is the easiest way to go).


Loosely, kimchi [1] consumption and exercise frequency are both (and independently) negatively correlated with social anxiety.

[1] As a stand-in for all fermented foods that contain or are likely to contain probiotics.


Are you pulling from a source for this? Care to link?


I'm TL;DR-ing The Fine Article.

Scroll up for the link.


For those with subject-matter expertise in probiotics: Is buying probiotic supplements a waste of time and money, and, if not, what are the strains one should be looking to purchase? Is there any known association between particular strains and particular effects (such as weight loss, or the anxiolytic effect described here, etc)?


Absolutely not a waste of time. I've experimented with all kinds of probiotics over the last decade.

In general, you want to buy high-quality probiotics, and finding the good ones isn't easy mainly because of the huge amount of astroturfing from manufacturers.

So here goes (note that these work for me, IANA doctor, YMMV yada yada):

VSL#3 and Therelac are the best broad-spectrum probiotics. Both have very high concentrations of bacteria, and require refrigeration. These are expensive, and typically used if you're trying to fix some condition. I make sure I always have a bottle of either stocked, and take it only if I feel I need to.

For day-to-day use, the following work great: Culturelle and Jarrow Formulas S. Boulardii. They're reasonably priced, don't require refrigeration, and the effects are noticeable in a day. I typically take a capsule of each twice a day (morning/night.)

Finally, make sure you get digestive enzyme supplements and take them religiously with meals that are hard on the digestive system (e.g., meat.) Twinlab Super Enzymes are what I typically get.

In general, the key effects I've noticed are: much improved digestion, lower anxiety, corrects GI issues, improvements in skin and oral health, and simply a better overall sense of well-being.

What's particularly cool is if you take them with foods that you love but typically upset you (burgers, pizza, other greasy stuff), extra doses of these probiotics right after the meals (along with the digestive enzymes) completely eliminate negative aftereffects.

Good probiotics are incredible, and I wish more doctors better understood them. Meanwhile, I highly recommend trying the above, experimenting (with good judgement), and finding what works for you.


It's quite possibly a waste if they were not refrigerated along the path from manufacturer to storage facility to store. iHerb ships refrigerated for good probiotics. Not sure about specific strains; I use Garden of Life Raw Probiotics, since they seem to have a good mix of different types and the manufacturer really makes an effort to keep them refrigerated.


I've tried Equilibrium, the Probiotic With The Least Searchable Name Possible.

http://www.generalbiotics.com/

It didn't seem to do anything at all for me, but I'm already fairly healthy.


It's mostly a waste. Very difficult to eat enough bacteria to have them compete with your extant gut flora.

This paper is questionable for other reasons.


Huh? What about Botulism and the bad E. Coli?


Both Botulism and E. Coli make you sick due to potent toxins produced by the bacteria, not due to them taking over your gut flora. The bacteria doesn't even need to be alive to make you sick if it's already colonized your food.


Good thoughts! Botulism is caused by a toxin produced by Clostridium botulinum.

E. coli is a little more complicated: some E. coli spp. produce toxins, and other species are enteroinvasive and actually can outcompete the gut flora (confusingly enough, other species of E. coli are also in the normal gut flora).

The main reason that some species of E. coli can make it to your gut and cause disease are that they have a variety of pathogenic factors that definitely would not be found in bacteria in 'probiotic' foods.

(and I see who you are replying to just edited their response to add 'bad' E. coli)


I just eat probiotic yoghurts like Actimel and Yakult. I've felt much better since taking up that habit, and it's nice and easy way to get started as they're available in most shops.


Also, margarine consumption causes divorce in Maine. More "true facts" at http://tylervigen.com/spurious-correlations.


I don't need an excuse to eat more kimchi, but hey, if it's helpful :-). Kidding aside, I love that more studies seem to keep coming out linking healthy diets with mental and psychological benefits.


This study doesn't do that, so much.

People who eat fermented foods have less anxiety, but we don't know if lower anxiety means people are more likely to eat fermented food or if eating fermented food reduces anxiety.


On a related note how do I find out what microbes are living in my sauerkraut, kimchi, kefir, or yogurt right now? There is no nutrition label.


Make your own! All 4 of those foods are surprisingly easy and cheap to ferment yourself. The best books on the subject are by Sandor Katz. There is lots of information online as well.

I drink homemade Kefir (with a bit of honey) every morning. Whether the effect is real or not, I feel better on the days I have it than the days I don't.


I chose Alex Lewin instead of Sandor Katz. Now I have this sauerkraut that smells pretty good, but I don't have the courage to eat it or the confidence that it isn't moldy. If I had chosen Katz would things be different?


To the best of my knowledge there isn't much you could do with that information - we don't really know what "good" or "bad" intestinal bacteria distributions are (besides obvious disease states), and variation between healthy individuals is tremendous.


The microbes are not an intentional additive in many foods, like sauerkraut. The 'wild' bacteria and yeasts naturally found on the ingredients carry out their reactions and proliferate, so the actual quantitative concentrations and strains can vary between batches.


Eating fermented food always decreases my social anxiety. But I think the hops and/or distillation process kills off any bacteria that may have been in them, so i have doubts as to the causal mechanism.


I see you. :)




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: