The Virtues of Spicy Food

I’ve been blogging recently about gut health, and emphasizing that an impaired gut often leads to systemic diseases (“Disease Begins in the Mucus,” June 23, 2015).

Impairment of the gut most commonly comes from (or is associated with) infections, or what amounts to the same thing, microbial overgrowth. Small intestinal bacterial overgrowth (SIBO) is a very common disorder, afflicting about 5% of younger adults and 15% of older adults. [1]

If SIBO is a precursor to systemic diseases, then any intervention that clears germs from the small intestine should help prevent disease.

Spices: A Natural and Beneficial Antibiotic

The traditional cooking spices, such as onions, garlic, and ginger, are all antimicrobial. [3] So if clearing microbes out of the small intestine is important for health, then eating spicy food should be a good thing. The hotter, perhaps, the better: the hottest spice, chili peppers, are known for their antimicrobial and antiviral properties. [4]

But how large are the benefits of including spices on food? Evidence toward an answer comes from a new study out of China [2] using the China Kadoorie Biobank database. The study asked 487,000 people at baseline, “During the past month, about how often did you eat hot spicy foods?” Then it tracked their health outcomes for 7 years and analyzed death rates as a function of the number of spicy meals consumed per week. In this case, “hot spicy foods” means meals with chili peppers or chili oil.

Interestingly, the mortality benefits of spicing food were large. Here’s a table showing the relative risk of dying from various causes, as a function of chili pepper consumption:

  No spicy meals Spices eaten 1-2 days per week Spices eaten 3-5 days per week Spices eaten 6-7 days per week
Death from all causes 1.00 0.90 0.85 0.83
Death from cancer 1.00 0.96 1.01 0.95
Death from heart disease 1.00 0.83 0.75 0.73
Death from diabetes 1.00 0.77 0.46 0.59
Death from respiratory disease 1.00 0.60 0.59 0.57
Death from infections 1.00 0.87 0.68 0.71
Death from all other causes 1.00 0.90 0.89 0.81

The biggest benefits were for diabetes and respiratory disease, then infectious disease and heart disease. This makes sense:

  • Diabetes is probably often caused by small intestinal infections that spread to the pancreas.
  • Respiratory pathogens infect the mouth and throat, where food spices can kill them; also, they are frequently swallowed, after which they can infect the gut as well.
  • Systemic diseases often enter the body through the gut, so anything that clears pathogens from the gut will reduce infectious disease.
  • Heart disease is also known to have a major infectious component.

In the subgroup analyses, there were some interesting correlations. Here’s Figure 2 from the paper:


The interesting thing here is that alcohol consumption eliminates the benefits of chili pepper consumption. Alcohol intake is a known cause of SIBO [1], and possibly the alcohol is removing the beneficial effects of the chili peppers.

Another interesting finding was that fresh chili pepper seemed to be more protective than dried chili pepper or chili oil. Probably there are more bioactive and antimicrobial compounds in the fresh peppers.

Neurotoxicity: Pro or Con?

The study authors note that capsaicin in chili peppers has complex biological effects. So it’s possible that not all the benefits of chili peppers were due to antimicrobial action in the small intestine.

Razib Khan, tongue in cheek, suggests another possibility: maybe chili peppers kill off the neurons of the gut. Capsaicin is a ligand for the pain receptor TRPV1 on neurons; this is why eating hot chilis can be painful. In high doses, capsaicin can actually kill neurons. [5]

Now, it happens that TRPV1 knockout mice live 14% longer than wild type mice [6]. If capsaicin could kill off the TRPV1-positive neurons, then maybe the effect on longevity would be the same as in the knockout mice, who have the neurons but not TRPV1.

Entertaining as this idea is, it would require massive chili pepper consumption. Rats lost neurons after intraperiotoneal injection of 125 mg capsaicin per kg body weight, equivalent to 10 g in humans [5]. Since pure capsaicin has a rating of 16,000,000 on the Scoville scale, ten times hotter than the hottest known pepper, this would be like taking 100 grams of the Carolina Reaper, 600 grams per day of habanero chilis, 6000 grams of serrano peppers, or 60,000 grams of jalapeño peppers. I don’t know about you but I would rather not eat a hundred pounds of jalapeño peppers, at least not at one sitting.

Another consideration is that the chili consumption would have to be sustained. Capsaicin can kill neurons, but if you stop eating the chilis, the neurons come back. [5] [7]

More fundamentally, the neurons of the gut are crucial for digestion, gut motility, and tissue healing. Losing those neurons would itself be a potential cause of SIBO and all the diseases against which chili peppers are protective.


Killing off neurons seems an unlikely path to health; nor is it likely a good thing to make your soup so spicy that it burns a hole in your stomach, as happened to one Chinese man.

The best course, it appears, is to eat chili peppers, but in moderation. Nearly all the benefits are obtained with a couple of hot and spicy meals per week.

Reminder: The Perfect Health Retreat

The October Perfect Health Retreat, taking place October 10-17, still has spots remaining. If you’d like to learn PHD home cooking from Shou-Ching, the science of good health from me, and a course in movement and relaxation from our star trainers Court Wing and Jae Chung, plus health coaching from me, while enjoying a delightful time on a magnificent beach and luxury property with hot tubs and pools, come to the retreat!

For more information, see our web pages or write me at


[1] Dukowicz AC, Lacy BE, Levine GM. Small intestinal bacterial overgrowth: a comprehensive review. Gastroenterol Hepatol (N Y). 2007 Feb;3(2):112-22.

[2] Lv J et al. Consumption of spicy foods and total and cause specific mortality: population based cohort study. BMJ 2015;351:h3942. Full text:

[3] Sherman PW, Billing J. Darwinian Gastronomy: Why We Use Spices. BioScience 49(6) June 1999: 453-463.

[4] Khan FA, Mahmood T, Ali M, Saeed A, Maalik A. Pharmacological importance of an ethnobotanical plant: Capsicum annuum L. Nat Prod Res. 2014;28(16):1267-74.

[5] Czaja K, Burns GA, Ritter RC. Capsaicin-induced neuronal death and proliferation of the primary sensory neurons located in the nodose ganglia of adult rats. Neuroscience. 2008 Jun 23;154(2):621-30.

[6] Riera CE et al. TRPV1 pain receptors regulate longevity and metabolism by neuropeptide signaling. Cell. 2014 May 22;157(5):1023-36.

[7] Gallaher ZR, Johnston ST, Czaja K. Neural proliferation in the dorsal root ganglia of the adult rat following capsaicin-induced neuronal death. J Comp Neurol. 2014 Oct 1;522(14):3295-307.


Leave a comment ?


  1. Not knowing whether they tested for this or not, first thing that came to mind was, those who prefer spicy food also shun sweets and sweet food. The benefit could well be just from eating less sugar, There are those that prefer not to burn their tongues, intestines and sphincters, except to appease the teasing of those very mucho macho, smoke your tasting buds, strangely chile-loving crazies, yet still not eat a lot of sugar either.

    • Maybe, but Chinese don’t eat a lot of sugar, at least historically they haven’t. This study compared Chinese to Chinese.

    • Carlos, where is this idea from? I’ve never heard it, and it definitely doesn’t apply to me. Do you mean that historically, cultures that eat spicy food don’t like sweets? Or do you mean that people who biologically are attracted to spicy food aren’t attracted to sweet?

  2. Paul, I’m curious: how do we square the benefits of spicy food with the benefits of fermentable fiber and probiotics? Do spicy foods act just on microbes in the small intestine before absorption or degradation before reaching the colon?

  3. On a related note, I recently discovered how excellent mortar & pestles are at extracting the most from fresh herbs. Highly recommended.

    Take a look at what results from Jamie Oliver mixing basil and olive oil:
    “Almost like chlorophyll.”

  4. “Diabetes is probably often caused by small intestinal infections that spread to the pancreas.” Could you please elaborate? I have never heard of that. Thanks!

  5. Saw on another website:

    Deeply flawed study, since consumption of spicy food in China follows geographic regions, and hence genetic populations. The folks who eat the spicier food may live longer due to alleles they were born with because of where they come from.

    • Hi Becky,

      The effect of genetics on population-level mortality rates is generally too small to explain results of the magnitude found in this study.

      But that doesn’t mean there isn’t some non-genetic factor which differs between regions. (See my comment below…)


  6. Hi Paul,

    My understanding is that traditionally, the spiciest food in China comes from Sichuan, Hunan, and nearby provinces. But the Sichuan Basin is a major rice-growing area, so these locations use rice as their staple grain. In contrast, most of, say, northern China both has (comparatively) less spicy food, and uses wheat as a staple grain.

    Wheat is associated in the epidemiological data with cardiac mortality, but not with cancer mortality — which resembles the pattern found for spicy food…


    • Hi Eric,

      Yes. But, cardiovascular disease more than cancer is a disease caused by chronic infections, so that difference is consistent with the antimicrobial function of spices. Moreover, spices reduce all-cause mortality, but it’s less clear that rice does so as well (see eg “Rice intake was inversely related to CVD mortality. There was no association between rice intake and cancer- or all-cause mortality”).

      It’s always hard to separate the confounders in these epidemiological studies. But with large effect sizes, there’s probably something real here.

  7. Hi Paul, I love that peppers are healthy, and as a migraine sufferer, they are sometimes recommended to relieve symptoms. But I have SIBO, rosacea, issues with controlling heat, especially on my face, and my current doctor has not ruled out auto immune disease. My TSH IS 4.5 and he’s not treating my thyroid (yet), but is focused on eradicating the SIBO and leaky gut. ( I’ve tested negative for antibodies, but I don’t know if false negatives are possible with Hashi’s.) He is suggesting that I try an AIP paleo diet for a month or two to see if any of my symptoms are improved, and then we’re going to add in Rifaxamin and Neomycin. I’ve followed PHD for about 18 mos.
    Two questions:
    1) what do you think of me trying an AIP version of your diet (either with or without egg yolks)?
    2) I don’t currently eat hot peppers or anything spicy because they light up my face (not just a vanity issue, but discomfort too). If I wanted to though, just a couple of days per week as you suggest, I see from your research that I would be getting some microbe die off benefits (including archaea? I have the methane type) especially if they’re fresh. But how does that impact the AIP trial?
    Thanks for your ongoing knowledge sharing and generosity!

    • Hi Susan,

      First, your doctor should be treating your hypothyroidism. The gut and face would heal better and immunity would function better if you had normal levels of thyroid hormone.

      AIP — I have mixed feelings. It is good for symptomatic relief in people who have a lot of food sensitivities, but tends to be overly restrictive and it can easily make things worse if it doesn’t provide complete nourishment. Generally, I would test individual changes to see if you notice symptomatic relief. AIP is a good resource for things to try. But think of the testing as diagnostic, not as a permanent therapy. For example, if you find that high histamine foods are a problem, reduce them for a time, but focus on improving histamine clearance with egg yolks, liver, choline, B6, B12, copper, and molybdenum. Similarly, I would not cut out egg yolks, but would mix the yolks with coconut milk and food and heat/cook them, making them more digestible, while discarding the whites. I would test nightshades, something like spicy peppers can be helpful if you don’t have a sensitivity to them.

      Yes, I think one meal a week with a spicy pepper might do you good. Also other herbs and spices, and things that acidify the small intestine — vitamin C, vinegar, bile supports (taurine, glycine).

      Best, Paul

      • Paul,
        Thank you very much for that. I have seen several functional medicine doctors over the past 2 years and have suggested treating my thyroid to all of them; but they all chant the same mantra: “we need to fix the gut first and that my be why your thyroid is off” or some variation on that theme. I understand and don’t disagree with that concept, but I also see things the way that you do, namely that with impaired thyroid function, how can the gut or whole body heal (especially since high tend to be a high stress person who recently began menopause, so probably all of my hormones are out of whack, which puts additional strain on the thyroid)? But my current practioner said I could try an OTC natural thyroid (apparently it’s the same as Armour) and see if I get any symptomatic relief. So maybe I should try that, though probably either for 30 days before or after doing the antibiotics course, just so I’ll know what’s helping and what’s not.

        I’m also with you on AIP too — torn. I’ve never had any directly traceable reactions to any foods (unless they cause heat, like too thermally or spicy hot, and alcohol of any type), and I’ve done several elimination experiments over the past 2 years, including with histamines, all without any “aha” moments. I even tried experiments with DAO enzyme, digestive enzymes, Betaine HCL — all with zero cost or benefit (other than the monetary and nuisance costs). As far as I can tell, nothing is a problem — nightshades, eggs including whites (although I’ve already cut back on whites based on your general recommendations), nuts, seeds, safe grains and occasional pseudo grains, etc. The only question mark still is dairy. I’m eating it in moderation; sometimes it can be a bit constipating, especially yogurt.

        Maybe I should add Molybdenum — I’ve never taken that. I’m compound hetero for MTHFR mutations, so I take methylated B12 and a bunch of other B compound stuff from Seeking Health, including their Liver Nutrients. I “try” to eat liver 4 oz per wk. (It’s more that I hate cooking it than I do eating it!) I eat nuts and chocolate and all-pastured beef and lamb daily, so I don’t supplement copper (unless you think that I should). Selenium and more copper from 3 brazils daily. I eat lots of fresh citrus juice as dressings for all of my food each day – primarily lemon and lime, but rarely vinegar. I take Vit C and Lysine 2-3 x per day, and also Vitamin D3 and Glycine daily. I take Taurine and E once weekly.

        I’ll try your yolk prep, though I’m not a big soup person, so I need to do some experimenting. Maybe an egg custard for breakfast or dessert! And, I’ll try to incorporate a spicy pepper into a meal 1-2/wk. I sometimes use jalapenos when I can tell they’re not hot yet, but it sounds like hotter is better for this purpose. I eat lots of non-hot peppers too, mainly red, yellow and orange.

        Thanks again Paul. You are so kind and generous with your time, intellect and heart.

        • Susan, re dairy and constipation, try only high fat dairy such as sour cream , butter and ghee. The high protein dairy foods can be constipating for some people

          • Hi Ellen,
            Yes, that’s what I do, except that I try to eat modest amounts of cultured products for the benefits too (and because I like them).
            Absolutely, I do find that high- or all-fat dairy makes me feel less bloat-y. I also think that I might tolerate raw better than pasteurized (I get full fat raw cow and goat milk direct from local ranchers who I know and trust). And almost all of my cheeses and cream are raw.
            But I’m too lazy (and/or inept/inexperienced) to culture the cream or milk, so I end up purchasing sour cream and whole milk yogurt and kefir. At some point I’d like to learn how to culture and try them raw to see if that makes a difference.
            I mainly use dairy they way that Paul suggests, as a pleasure food/condiment, in coffee, and not as a mainstay. So I really only notice an issue if I over-indulge and drink kefir or eat yogurt straight, rather than as just a dollop or two in dressing, etc.
            Oh, and my raw milk is all-Jersey and the goat is a naturally high-fat Golden Guernsey; I get mostly raw goat, sheep or Jersey cheese — all are A2 casein.
            Thanks again!

  8. Hi Paul,

    Thank you for another excellent article. It is kind of ironic how alcohol, which is more antimicrobial (or anti-life) than any other substance is bad whereas the spicy food is good for these exact reasons!

    I was wondering why you singled out spicy foods and not just went with spices, most spices such as ginger and coriander are very antimicrobial as well.

    Best, Palva

    • Hi Palva,

      I didn’t single out spicy peppers, the paper did. Antimicrobial spices like onion, garlic, and ginger are probably just as good for you.

      Alcohol does have some health benefits in low doses and it may be that its antimicrobial action is part of them. But too much alcohol is not good for the gut.

  9. Tedenski pregled #14 - pingback on August 13, 2015 at 3:18 am
  10. “Respiratory pathogens infect the mouth and throat, where food spices can kill them; also, they are frequently swallowed, after which they can infect the gut as well.”

    Paul, none of this makes sense to me. You are suggesting respiratory infections spread to the rest of the body. I feel like the opposite is far more common. Also, if you are dying *from* respiratory disease, this doesn’t make sense.

    My guess would be that either the spiciness changes some part of the immune system that is more critical in the lungs than other parts of the immune system (e.g. glutathione).

    Also, maybe some metabolite(s) of the spicy food are exhausted through the lungs (like alcohol) and directly expose the lungs to those molecules.

  11. Latest in Paleo 140: Killer Kale Theory – Humans Are Not Broken - pingback on August 16, 2015 at 10:30 pm
  12. I look through things these days with an acid reflux filter. I don’t think spicy foods would be good for someone with some kind of gerd or gastritis.

    • As a former gerd/reflux sufferer, I feel that spicy food has little or no bearing on the situation. For me it was primarily wheat or some combination of caffeine, simple carbs, and stress. I recommend Peptiguard from Thorne – this along with another Thorne product were prescribed by my holistic doctor and they eradicated the reflux over time. Every so often, if I cheat too badly, I’ll get heartburn and Peptiguard helps me during that time.

  13. This is making me feel SO much better about my raging kimchi addiction 🙂

  14. Hi Paul,

    More generally: What are the most important beneficial natural antimicrobials that should be consumed regularly? My guess is such a list might start:

    Hot peppers.
    Onion-family spices: Onion, garlic, etc.
    Anthocyanin-containing foods: blueberries, blackberries, etc.
    Proanthocyanidin-containing foods: cranberries, strawberries, etc.
    Vinegar and other organic acids (for biofilms).

    Do these all deserve the “most-important” label? And what else does? Possible candidates include:

    Coffee or tea.
    Black pepper (efflux pump inhibitor).
    Other antimicrobial spices: Ginger, allspice, cinnamon, cloves, bay leaf, vanilla, oregano, thyme, marjoram, rosemary, cumin, lemon grass, tarragon, etc.
    Other colorful fruits and vegetables: beets, tomatoes, citrus, etc.

    But I don’t know if any of these belong in the first group, or whether they belong in a “second tier”. In particular:

    – Coffee and tea are fasting-friendly, unlike everything else. Is this important? If so, which one is most likely to be beneficial?

    – Black pepper is an efflux pump inhibitor [3], increasing toxicity of other compounds towards both pathogens and humans. How beneficial is this?

    Many of the antimicrobial compounds here are polyphenols, so I thought the following might be relevant [2]:

    “Among the polyphenols, isoflavones are known to be more bioavailable followed by phenolic acids, flavanols, flavanones, flavonols and lowest bioavailability was seen for anthocyanins and proanthocyanidins.”

    This is interesting because:

    – The polyphenols with the highest bioavailability (isoflavones) are found primarily in legumes (soy and peanut), which face a selective pressure to be toxic towards mammals. Indeed, isoflavones are endocrine disruptors.

    – The polyphenols with the lowest bioavailability (anthocyanins and proanthocyanidins) are the primary polyphenols found in berries, which face a selective pressure to be nourishing towards mammals. They appear to be quite beneficial.

    – The polyphenols with intermediate bioavailability are found primarily in nuts, seeds (of non-leguminous plants), tea, and chocolate, which don’t have a strong selective pressure either way.

    Most of the remaining antimicrobial compounds are mono- or di- phenols, for which I couldn’t find great references on bioavailability. But on general grounds, the key factor should be the number of rotatable bonds, with higher values corresponding to lower bioavailability (Veber’s rule [1]). There are 9 rotatable bonds for capsaicin, 8 for curcumin, and <= 4 for every other natural antimicrobial mono- or di- phenol that I am aware of. So I'd guess capsaicin and curcumin have the lowest bioavailability, and thus that hot peppers and turmeric are among the most beneficial spices (in terms of the ratio (toxicity to pathogens) : (toxicity to humans)).

    Remaining antimicrobial compounds include:

    Antimicrobial peptides in animal foods such as milk and honey, which definitely look safe; although honey has lots of fructose which could promote bacterial growth, and casein could interfere with motility.

    Organosulfur compounds in onions and garlic, which you indicated might even be beneficial when systemically absorbed (e.g. item #5 from






  15. The relationship with alcohol intake is fascinating.

    I wonder if capsaicin downregulates or blocks alcohol dehydrogenase (ADH)? The damage done by methanol is directly related to ADH levels, with higher levels clearing ethanol more quickly, then causing more rapid rates of formaldehyde production from methanol.

    There is a relationship between ADH levels and alcohol intake: for a man, one drink a day is optimal, because it isn’t enough to upregulate ADH levels, and the extra ethanol reduces conversion of methanol to formaldehyde, thus reducing damage. Somewhere north of 1 drink/day will upregulate ADH levels, thus explaining the much higher risks of binge drinking (these people have upregulated ADH, but don’t have enough regular ethanol intake to prevent methanol toxicity, the worst of all worlds). Ethanol is also toxic, but at the 1/2 to 1 drink per day level it dramatically reduces methanol toxicity, without adding a lot of toxicity itself, so that there is a net longevity/health benefit.

    We don’t have enough information from this study to know what’s going on re alcohol intake, unfortunate that they didn’t provide us with more information.

    @paul I hope you will read Woodrow Monte’s superb book, While Science Sleeps. He’s found a big piece of the health puzzle: not just aspartame and smoking, but our intake of higher-than-traditional levels of methanol.

Leave a Comment

NOTE - You can use these HTML tags and attributes:
<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

Trackbacks and Pingbacks: