The elixir broccoli of life

New papers in The Lancet confirm that aspirin helps prevent cancer, as well as heart attacks, which has been known for a long time.
These are very, very strong findings. One of the papers (Rothwell et al – has sidebar links to the others) is a meta-analysis of 51 controlled trials involving 77,000 participants. The p-values of the effects (depending on what they were looking at: 3 vs. 5 years, incidence vs. deaths, or women vs. men) range from 0·01 to 0·0003. The latter means there is a 1 in 3,300 chance of observing these data if there’s really no connection, getting into physics territory. The size of the cancer-reducing effect for these metrics lies between 13% and 24% – not a miracle but still substantial. Rothwell doesn’t offer calculations of increased life expectancy; but the 5-year effect was stronger than the 3-year one, so aspirin is not just delaying onset by a year or two.

At this point I assume that further controlled trials become unethical, except to determine optimum dosages.

Big Pharma will not be putting a lot of effort into marketing its off-patent Victorian wonder drug, so takeup will depend on word-of-mouth between consumers as much as on medical advice.

Should you go out and start taking baby aspirin? A year’s supply will set you back all of $25. There is a heightened risk of ulcers and intestinal bleeding, and a slight one of stroke. If you have a specific vulnerability to cancer – like Lynch syndrome – it looks a no-brainer, and still a pretty good deal if you don’t. Don’t take my word for it, and as they say it’s safest to seek medical advice. However, researchers on the subject take aspirin themselves.

Most of us die of either cancer or heart disease. Since aspirin is good against both, it’s starting to look like a longevity pill.

Some are even calling it ‘vitamin S’: a needed component of our diet, rather than a medicine. The argument goes like this. Plants, including the ones we eat as vegetables, naturally produce aspirin-like compounds, salicylates. as defence against their pathogens. In the ancestral environment our diet naturally had a lot of these chemicals. Today we don’t eat enough leaf vegetables anyway – this is standard nutrition advice. Second, there’s a nice theory that the vegetables we do eat are far too perfect. Modern supermarket veg is grown in near-sterile conditions; pathogens are kept at bay with pesticides; and we or the supermarkets throw away blighted leaves or heads.

Fig.1 Unhealthy broccoli

Source

The basis for this imperfection claim doesn’t look very strong. The articles I’ve seen cited are by John Paterson, a Scottish medic, who found much higher levels of salicylates in the blood of vegetarian Buddhist monks (2001, Journal of Clinical Pathology, vol 54, p 553) and organic soups (2002, European Journal of Nutrition, vol 40, p 289), compared I suppose to ordinary haggis-fed Scots blood and supermarket soups. Nobody seems to have done a controlled trial in a greenhouse. (I mean, of vegetables with and without diseases, not of monks, Scots or soup.) However, I’ll go with Paterson for now because the story makes sense.

So Republicans be warned. If Obamacare is held constitutional, liberals are not only going to legislate that you must consume Interstate-Commerce, Necessary-and-Proper broccoli. For your health and the General Welfare, the Life Panels will force you to eat blighted broccoli.

Fig.2 Healthy broccoli

Source

Comments

  1. Brett Bellmore says

    The interesting thing about this is that aspirin has *so many* different effects, that it’s widely considered that, were it invented today, you’d never be able to get it past the FDA. It would fail trials on its tendency to cause gastric problems alone.

    • says

      For headaches perhaps, but as a cancer preventative? Cancer treatment drugs have horrific side-effects, but are approved anyway because of the alternative.

    • Warren Terra says

      Do you have a cite for your claim that your position is “widely considered” to be plausible?
      After all, I’m not sure you’re even right about headaches – I believe (though I could easily be wrong) that the side effects are associated with chronic use, and aspirin, like many over-the-counter painkillers, is only meant for acute use. Indeed, the whole story of the development of COX-2 inhibitors (Celebrex, etc.) was that they would in theory affect the pain-relevant target of aspirin while not causing the known off-target effects (gastric problems, for example), and thus their developers hoped that they could be prescribed for sustained use to fight pain (arthritis especially) in a way that aspirin and other painkillers could not.

      • Anderson says

        Do you have a cite for your claim that your position is “widely considered” to be plausible?

        Warren, you’re talking to Brett.

      • Brett Bellmore says

        Try, for instance, Aspirin: Not Approvable. Or No Refills. Or, (echoing James’ own remark about the ethics of a proper trial) Fdea panel says no to asprin.

        People have been saying things like this for literally decades. The only reason we’ve got aspirin is that it’s grandfathered in, if invented today it would, at best, be a prescription only drug.

        And I’m betting caffeine today would be at least schedule 2, if it were freshly invented.

    • Bruce Wilder says

      Just noticing how Brett constructs an argument by placing a counterfactual, where one might expect a fact. He bolsters the credibility of this pseudo-fact with a confident assertion, “that it’s widely considered that . . . ”

      I see another commenter, drawn into a trance I guess, asked for a cite — like there could be a cite to confirm the factualness of a counterfactual. I guess someone could, conceivably, marshal actual facts in support of a theory, proposed in the form of a counterfactual, and that could be informative.

      I did the requisite Google, of course, and found quite quickly an essay from the right-wing libertarian Manhattan Institute, using this assertion as an introductory, rhetorical counterintuitive hook. It reminds us that aspirin has side-effects. And, paracetamol (acetaminophen) does, too, and penicillin. The peroration in this essay:

      What are we to make of all this? It’s possible to be both glad and worried. We can be relieved that we’ve learned so much more about pharmacology, ensuring that the drugs that manage to gain approval today are . . . safer than ever. Or we can think about how people seem to use aspirin and the other legacy drugs anyway, safety problems and all, and wonder how many more useful medicines we’re losing by insisting on a higher bar.

      I definitely see the point of the former, but I lean a bit toward the latter. Drugs have side effects, after all. Imagining that the older ones didn’t, or pretending that the newer ones mustn’t, is a recipe for disappointment. We’re better served by reality, as it was and as it is.

      I loved that last: “We’re better served by reality, as it was and as it is.” Good advice, that.

  2. marcel says

    Brett wrote: It would fail trials on its tendency to cause gastric problems alone.

    Beans and garlic too, no doubt. Perhaps even dairy products because of the many who are lactose intolerant.

  3. says

    Heck, straight water’s toxic dose is hardly that far from the therapeutic one. But no, if aspirin were introduced today, it would be for some niggling little disease, and then all the other uses would be off label. If it takes years to get drugs off the market that flat-out kill people under normal use, neo-aspirin would be around for a while.

  4. yoyo says

    Note that one of the reasons we need aspirin to block COX enzyme is the grossly excessive intake of the substrate for that enzyme – polyunsaturated omega 6 fat from seed oils and meat.

  5. Daniel says

    If calabrese (as in the picture) produces salicylates to protect itself, why would an “unhealthy-looking” leaf or head have more salicylates than a “healthy-looking” one? So I don’t think the health police will be insisting on folk eating blighted leaves, the unblighted will be just as good.

    And I think there is some confusion over pesticides and pathogens. With different, and correct, words, then the argument is an interesting one – let the plants fight their attackers a bit more, and they would be better for us to eat. Certainly organic philosophy is to feed the soil, then the plants grow strong, then they can stand up better for themselves. Oh, and then they will be better for us to eat too.

  6. Brett Bellmore says

    For the same reason humans, who produce pigments to protect themselves, get darker in response to UV: Biology is cheap; You don’t produce protective substances in large quantities if you don’t need them.

  7. Andrew Laurence says

    Where do you shop that a year’s supply of baby aspirin costs $25? I buy the Kirkland brand at Costco. It’s designed to dissolve in the small intestine, not the stomach, so there’s no risk of gastric distress, and I think a six-month supply costs around $3-4.

    • James Wimberley says

      Quick Google. The $25 was for Bayer. I’m glad it’s only an upper bound!
      I’ve just paid €1.36 at my local Spanish pharmacy for 20 (Sanofi-Aventis), which makes €25 a year. Spanish pharmacists are protected by restrictions on supermarkets, which can’t sell OTC drugs. UK prices are similar: £20 at Boots (per website).

  8. Brad says

    James,

    I’m curious as to the logic to reporting the point on the distribution associated with the results (after the test) rather than the alpha level set prior to data collection – I assume that there was an apriori alpha level (probably .01). Since the fact that the P was estimated implies some random variation, where exactly a given set of observations lie on the distribution is not really what should be reported – that is still a randomn result. (And yes, I know that it is commonly done, but…)

    I’m not suggesting that these results are in any way flawed – they are extraordinary, I’m only questioning the way they are reported.

    • James Wimberley says

      I think your point is better addressed to the authors than to my short pop report. The full article is behind a paywall so I was going by their own summary, which omits the experimental designs of the 51 trials and the technique of aggregation of the meta-analysis.
      When credibly qualified authors publish, in a top peer-reviewed journal, unequivocal findings of very high statistical significance, based on an enormous sample, and consistent with other established science, I reckon we have to take the findings as sound. At least until somebody else in the trade picks large peer-reviewed holes in their work. Wanna bet?

  9. NCG says

    Can I wait until I’m 60 to start taking it? When do most people get cancer? I don’t smoke, but I do breathe dirty air.

    • says

      One good way to shorten your life expectancy is to get your medical advice from general-interest blogs. I posted this on the basis that my readers,if interested, would either read up themselves or (as I recommended) talk to their doctor.

  10. Brett Bellmore says

    I’ve had a round half dozen attempts to give Warren his cite go to comment limbo. Let him use google himself.

    • Warren Terra says

      I didn’t spend a lot of time reading around. I found an essay by Derek Lowe, whose blog I’ve read and whose judgment I respect. He expresses concern about aspirin, but mostly in the context of a wider point that I’m not fully convinced by, in which he asserts in essence that rare extreme reactions (allergies, for instance) kill forward progress on drugs that are beneficial to most under controlled circumstances. Maybe he’s right; he’s certainly better informed than I am. But given that aspirin in fact does not seem to be causing major problems when used as directed, this seems like more a commentary on the incentives in drug design, and the need for better assessment of risk data.

      • Brett Bellmore says

        Warren, that essay is one of the links I *tried* to supply, so I’m glad you found your way to it. Not sure what was going on, but every comment including a link, even a website filled in on the comment form, got intercepted.

        “But given that aspirin in fact does not seem to be causing major problems when used as directed, this seems like more a commentary on the incentives in drug design, and the need for better assessment of risk data.”

        A lot of those incentives are based on what it takes to get past the FDA… We, the end users, do not consider the averse reaction rate of something like aspirin to be excessive. That doesn’t mean the regulators agree.

        I am personally persuaded that we should simply abolish the FDA. It is significantly more restrictive than most countries’ equivalents, without being measurably more successful, and having a disturbing record of attempts to violate the 1st amendment. (Ironically, especially on the subject of… Aspirin!) We could save considerable money, and have better results, by simply adopting a rule to the effect that any drug approved of in a list of a half dozen or so 1st world countries was automatically approved for use in the US.