A Blog by

Resveratrol Redux, Or: Should I Just Stop Writing About Health?

(Update, 5/13: This post generated a lot of discussion on Twitter, which you can see Storify-ed here. I also talked about these issues on NPR’s On the Media, which you can listen to here.)

The science of health is so, so confusing, I almost wonder if it wouldn’t be better for journalists to stop writing about health altogether. Or at least to dramatically change the way we do it.

Take one of the biggest health stories of the last decade: resveratrol, a compound found in certain red wines that has been shown to extend lifespan and/or curb disease in yeast, fruit flies, fish, worms, and mice.

Searching the New York Times archives for “resveratrol” gets you 156 items. Here’s a sampling of the headlines:

August 2003: Life-Extending Chemical Is Found in Certain Red Wines

November 2006: Yes, Red Wine Holds Answer. Check Dosage.

January 2011: Doubt on Anti-Aging Molecule as Drug Trial Stops

August 2011: Longer Lives for Obese Mice, With Hope for Humans of All Sizes

January 2012: University Suspects Fraud by a Researcher Who Studied Red Wine

November 2012: Resveratrol Ineffective in Normal-Weight Women

March 2013: New Optimism on Resveratrol

Is your head spinning yet?

From what I can tell, there’s nothing overtly wrong with the journalism in any of these stories. Most are based on a new study (or studies), and include varied perspectives of scientists who had nothing to do with the research. The reason the stories contradict each other is because the studies contradict each other.

This happens in science all the time; it’s even supposed to happen. Think of all those models of the atom you learned in chemistry class: from Thomson’s plum pudding to Rutherford’s nucleus to Bohr’s energy orbits to Pauli’s electron spin. Two steps forward, one step back, science moves along.

But when it comes to writing health stories, it’s hard — really, really hard — to include that slow scientific progression in a way that a reader will absorb. And I think that’s because readers don’t seek out health stories to satisfy abstract intellectual curiosities. They want to glean some kind of practical knowledge. How can I avoid sickness / lose weight / feel better / live longer?

For some messy health issues — such as whether it’s dangerous to drink while pregnant, say, or whether to get screened for cancer — the stakes are high. Resveratrol is not as serious. For most people, drinking a glass of wine or taking a daily resveratrol supplement is not going to do any biological harm. But there are other kinds of harm. Searching amazon.com for “resveratrol” gets you 2,186 health and personal care items, including supplements costing dozens or even hundreds of dollars.

I got thinking about this because of a study on resveratrol that came out today in a solid medical journal, JAMA Internal Medicine. Fifteen years ago, researchers collected urine samples from 783 older people who live in the Chianti region of Italy, where drinking red wine is common. It turns out that the level of resveratrol in the participants’ urine could not predict anything about their health outcomes. Those with the highest levels were just as likely to have inflammatory markers in their blood, and just as likely to get heart disease, cancer, and to die.

So I read that study and thought, this is important: My readers who buy or are thinking of buying resveratrol might appreciate knowing that its benefits haven’t panned out in people, at least not yet. Sure, a future study in people might report some benefit of resveratrol, but for now all I can do is offer the current state of knowledge. And that’s better than nothing, right?

But then…maybe it’s not. Take a look at those headlines again. I suspect a general reader is not coming away from those saying, “Gee whiz, look at the long and bumpy road to scientific progress!” They’re more likely to be saying, “When will those scientists get their act together?” Or worse, “Why do we keep dumping money into this capricious discipline?”

I don’t have any grand solution to this. I’ll undoubtedly keep covering health stories, because I believe in the public’s right to accurate information. And I believe in the process of science, however slow, to ultimately figure things out.

Still, is there a way that journalists could do this better?  How should I have covered the latest resveratrol study? Should we switch to a more explanatory, wiki-like model, so that a single study’s results are more fully contextualized? Should we be writing stories about batches of studies — maybe the last 10 studies of resveratrol, as opposed to the single newest one? Are headlines the real problem?

If you have any preferences or suggestions I’d love to hear them. I’m not likely to change the Health Journalism machine, but I’m more than happy to experiment on this blog.

44 thoughts on “Resveratrol Redux, Or: Should I Just Stop Writing About Health?

  1. The problem is bigger than science: it’s religion. Denial of death. Belief in the supernatural; magical thinking. TOO MANY INSTRUCTIONS. You’re going to die anyway. No one goes to heaven. Just live!

  2. I don’t pay much attention to science reporters who lay out 5Ws and call it good. I follow science writers who put the latest development in context. Blogs are better than set pieces because the writer can tell the story as it develops or as his or her understanding sharpens. Science is about successive approximations; the same goes for science writing.

  3. The journalism is fed by the Press Release machine. The Press Release will be written by someone with more zeal to boost their institution than in-depth knowledge of the subject. Most journalists seem to lack the time, skill, and knowledge to drill down to the original report, which as like as not is a conference abstract lacking the detail needed to evaluate. Googling just finds you what’s been looked at most often, by readers looking for miracles.

    As it happens, I *like* red wine anyway.

  4. Resveratrol significantly improved mental function. 32.8% of those with the lowest resveratrol excretion levels scored 24 or less on a mental tests, versus just 16.4% among those individuals with the highest resveratrol urine concentration. Should be big news! But Virginia Hughes chose to say “It turns out that the level of resveratrol in the participants’ urine could not predict anything about their health outcomes. ” That’s selective journalism for you!

  5. While I fully expect to die I would like to enjoy life while I live. Good health helps me do t
    hat. Nothing supernatural about living life to the fullest.

  6. I think we need more metastories about science and less stories about latest studies. I know I probably don’t represent the majority of the population in that regard, but I find science in itself is extremely interesting.

    For example, to counter the whole “this and that is healthy/unhealthy”-craze I think we need a lot more explanations why health science is pretty hard. Write about publication bias, write that studies without control groups are most likely not telling what they seem to tell, write about that stuff. It’s interesting.

  7. I do like your ‘batch’ idea, as no one study can give a feel for the consensus as any given time.

  8. I agree with Peter Kent that context is the important ingredient that is too often left out of scientific reporting. However, I disagree that there’s a blogs are necessarily better at this. One of the best ways of adding context to an article is to interview knowledgable sources – researchers working in the field who know the background, the caveats, the ins and outs of the methodology used. Traditional news outlets should always require that additional sources be contacted for comments, and good journalists will ensure that they have done this. Blogs have no such requirements, though many bloggers are just as rigorous in fact checking as journalists working in more traditional media outlets.

    One model of reporting that doesn’t necessarily consider context fully, but is nonetheless quite good at picking apart details of a study is the UK’s NHS choices. The have ‘behind the headlines’ reports that dissect studies that have been headline news (where did the story come from, what kind of research was involved, what were the results, how were they interpreted…). For example: http://www.nhs.uk/news/2014/04April/Pages/Could-a-carbohydrate-curb-calorific-cravings.aspx

  9. It bothers me when I keep seeing areas of science (or in this case, journalists reporting on the science) where people are jumping to conclusions being compared to things I learned in physics (or chemistry) class such as atomic models going “from Thomson’s plum pudding to Rutherford’s nucleus to Bohr’s energy orbits to Pauli’s electron spin.” As Peter Kent said, “successive approximations.” It was understood that these represented the atom to the best of our current knowledge. With reporting on health (how much is it the fault of the researchers? I don’t know), it’s largely a matter of making eye-catching headlines. Yes, a “life-extending chemical” was found — but read on and find it was only shown to extend the life of lower animals… and as another story pointed out, “check dosage.” And did anybody suspect Thomson, Rutherford, Bohr, or Pauli of fraud? Yet point to frauds and hoaxes in other areas of science and again we get this, “Oh, that’s just the way science goes…” Apparently some areas of science are allowing a lot of slack, speculation, sensationalism — and even excusing wrongdoing — as if those are normal parts of scientific investigation.

    And to “gina” and(?) “gina rex” who seem(s?) to have seen some reason to drag in a personal vendetta against religion: Take a chill pill, lady. Lots of people live long and very happy lives while following many “instructions,” and believing in the supernatural and life after death. Many also do many good things for others here in this world, and leave a legacy of joyful living to their children and grandchildren.
    I’m sure many atheists live similar lives, but I’d like to point out that being religious and all that isn’t a problem or handicap, even with an evolutionary measure. The folk I pity are those who convince themselves they are only really living because, and to the extent that, they use drugs, have as many wild sexual encounters as possible, perform dangerous stunts, and/or pat themselves on the back for “really living” more than other people — and then die young from an overdose, disease, or “accident” and too often don’t leave any legacy behind except bequeathing on others a similar fate.

  10. I think it would be responsible to require a certain level of statistical rigor and number of reproduction studies (that agree with each other) before reporting to the public things that will cause them to change their behavior, buy supplements, or seek medical help. We have this in science with p-values but it’s not enough. We also need replication. Unfortunately the initial studies are public (and should be) and there will always be unscrupulous journalists, press release writers, and supplement manufacturers who hype statistically (and reproductively) insignificant findings for their own gain.

    But a good start would be for journalists, scientists and statisticians to agree when it is appropriate to communicate findings to the public. How many reproduction studies with what combined p-value? The fact that reported results do not follow the expected statistics worsens this problem. See the Amgen replication study of 53 preclinical cancer drugs. So bias and fraud are creeping in even before publication…


  11. All of this:

    “Should we be writing stories about batches of studies — maybe the last 10 studies of resveratrol, as opposed to the single newest one? Are headlines the real problem?”

    When scientists write major articles that provide a narrative they typically do so in the form of a review of many studies over time. This is how journo’s should approach it. Usually. Game changing eurekas are rare. But hitting a point – a threshold – upon which the collective data over multiple studies provide a sufficiently robust picture to draw conclusions happens frequently enough. Journo’s need to learn to recognize when this has occurred. Which means they need to be reading, collecting, annotating, storing, and tracking science articles over time and evaluating data independently in addition to interviewing a bunch of people. You don’t have to be experts in the field, but you don’t necessarily need to be an expert to look at numbers once the meaning of those numbers has been explained to you. You often just need a basic grasp of statistics and the scientific method.

  12. I recommend checking out review journals like Nature Reviews, The Cochrane Collaboration or JAMA Clinical Evidence Synopsis to see if their approach can be better adapted to science journalism as opposed to the usual single-study-cum-university-press-release approach that I often see.

  13. One of the practices I noted and appreciated very early in Ed Yong’s original blog style was the foot note inclusion of prior blog, paper or other on-line relevant topics, allowing me further quick access to older information. Following with that idea, including not only your own previous blogs on a topic but a linked list, similar to what you show here, giving pros and cons of the latest “wonder” supplement would go a long way in helping the reader get the best current picture.

  14. David Bump: Religious people are the only “good” people. Anyone who is science-based is decadent, evil, uses drugs and is a sex addict. Like I said: religion is the problem. All this “study mania” is NOT SCIENCE, it’s about getting funding. It’s about profit. I’d like to see the funders who are behind the studies identified – Media contributes to ignorance about science: Dr. Oz is a pimp for useless products. It goes on and on…

  15. I think it’s all part of a broader issue where people really want/expect science to be definitive, when it basically never is. Within the scientific community, people are really skeptical of “experts” who say, “I’m 100% sure about this issue”, but outside of the community, the opposite is true. So I think a big part of reforming the machine is going to be about making people aware that science is SUPPOSED to flip-flop and overturn itself. That’s what makes it’s awesome. And it takes a long time to build a base of certainty that you can base practical advice on.

  16. I really like the batch approach and would love to see more of these kinds of stories. Often they take longer to report and write, and a bit more digging so they are perhaps not as financially effective. I wish that science journalists would not pursue the next new thing as much as we do, and write longer “this is what is going on and here’s a tiny bit more data to add” stories. These are the stories that really stand out.

  17. Journalistic ethics and rigour should be your guiding principles. So many people have been led astray by incorrectly reported articles, and worst still journalists allowing their personal ideology to get in the way. I can think of many tv journalists who are guilty of this. Your average Jo doesn’t know about fallacy and bias. We can’t all read a study and appreciate its nuances. We rely upon ethical honest journalism to help guide us through the ever – changing jungle of scientific research. Thanks for your efforts!

  18. I disagree that religion is the problem; I know a few brilliant scientists and resarchers who are also very spiritual in mode. Not only that, just because I am “religious” doesn’t mean that I believe that everying is “magical”; I suspect that many others who are scientifically minded but also spiritual feel the same way. While it may be a problem in some groups, I feel like that’s scapegoating, and scapegoating never solves anything.

    (Unless by “religion” you mean this newfangled new-age pseudoscience tripe, in which case, I entirely agree–that is dangerous.)

    I really feal, instead, that it’s laziness which is the problem. A lot of people seem to be too lazy to do the research, or read past the headline, and that’s going to be a major challenge to overcome, because most people, as you point out, just want to know the answer right now as it stands, and don’t want to wait for the progress of science, ALONG with the desire for sensationalized soundbites that bring funding to these studies, which of course, since people are people, they misappropriate and wave around like it’s the truth–which almost creates a pseudoscientific feel in many cases, and THAT worries me.

    Only when these issues are fixed, somehow, will we solve this problem. Dedicated science bloggers who take the time to discuss the issues in depth in an easily accesible manner which is easy to comprehend are a huge asset in this arena, I think–which is exactly why I read blogs like this.

  19. We need blogging and journalism on health now that Big Pharma are giving out manuals that mention only their products and not any alternatives. A health writer who is not a health professional must have background in chemistry, physics, biochemistry, biology, and pathology.

  20. Yet another 2nd, 3rd, or 4th hand account of the “latest” on red wine, and this one also puts the boot in to chocolate: http://www.heraldscotland.com/news/health/bin-the-bottle-of-red-and-chuck-the-dark-choccy-latest-research-refutes-health-benefits.1399952384?utm_source=headlines&utm_medium=email&utm_campaign=email%2Balert No reference to the actual publication, alas; just a soundbite from the lead researcher, who is himself hardly the best person to evaluate the significance of his finding or, in this case, non-finding.

  21. Someone – whose name I’ve unfortunately forgotten – at the scienceonline conference suggested what seemed to me like a very interesting approach to these issues: changing how we frame questions, to ask not ‘is red wine good or bad?’ but ‘what does red wine/resveratrol do inside your body?’. This allows the writer (and the reader) to explore why different studies might find different answers.

  22. Great post! It looks like Satire when you say I should stop writing on health 🙂 Well, health is wealth and you should write what your heart says! I usually write for the Reddit alternative https://tyger.ac and its my way of showing things Thanks!

  23. Good article, and right on target that general readers are looking for “actionable” guidance and not the latest scientific tidbit. Context and expert interview evaluating the “news” would help, but surely that’s true of any news. I assume the reason context is not provided routinely has to do with the pressure for speed. And re:headlines, I assume there is pressure to use the most sensational attention-grabber possible, unless perhaps the journalist/author/blogger is him/herself the reason the reader reads a story.

  24. Outstanding! For me, your piece naturally evokes lots of questions and provokes the curious mind. While I have certainly learned from your informed content, as a reader I can connect the various questions and challenges you raise to many different dimentions of our millennial life. Specifically: ethics; science; technology; privacy; civics; religion; politics; economy; fidelity just to name a few. My humble opinion is that it is tempting yet not terribly helpful to construct and communicate ideas by way of metaphor. It’s noisy and cluttering to my mind. Writing that is bound to rigorous, fact-based reasoning strikes me as far more powerful and potentially able to influence our world for the better. I look forward to your future elaborations hopefully on this challenge for science writers.

    Well done.

  25. To try to diminish the health value of Resveratrol based on one small scale, non placebo controlled, now double blinded trial of a wine that is not even known for having significant amounts of Resveratrol in it is pandering to the big pharmas who own or virtually own the media. A simple scholar google search will return over 100,000 studies, papers and investigations, including well designed human clinical trials which contradict this one study which was far too small to be meaningful, not double blinded, not placebo controlled, not randomised, and whose premise was flawed. One would not expect to find the metabolites of Resveratrol in the urine of wine drinkers. The sulphated and glucoronated forms of this compound have a half life of literally 12 minutes. It appears that big pharma is still determined to push their attack on resveratrol. Why this one very weak and hardly compelling study was given any press at all is the question when it is contradicted by literally thousands of far better designed, properly run, scientific studies. This study was funded by big pharma. What do you expect in terms of conclusions? It is irresponsible of this news media to publish it out of context ,and without mentioning the thousands of other studies that contradict it.

  26. Carl Sagan was a scientist with a great passion for science in general. He spent much of his life trying to explain the mysteries and wonders of science to general public (among many others before and after him). It has been a hard task always. Specially in this days, when many people has acces to internet (with all the good and bad that internet has). So, Virginia, good luck and never dismay: clinical studies are a mine field even for specialist but if you are capable of keeping track on novelties and explain it to the public…well, that will be succes. Actually I think is more than necesary as a public service.

  27. I also fully expect to die and I definitely want to enjoy life while I live. Red wine helps me do that, whether it contains restrewhatsit or not. That’s because it tastes nice and gets me pleasantly scoofed. I seriously don’t care if it contains vitamin Z, alien body secretions or fermented mice, and whether or not it will cause/prevent/have no effect on heart disease/cancer/beri-beri is a sweet irrelevance to me. Red wine is good.

    As for “health” reporting all I ask of journalists is that you stop trying to make me feel guilty about my little pleasures, because quite frankly you’re wasting ink.

  28. “you should write what your heart says!”


    Nope, boring, sorry.

  29. For those of us who actually attempt to keep current with these types of studies, hoping to forestall illness or at least to slow the progression of current maladies, the exasperation when yet another of these research “reversals” is published is so disheartening as to make one consider just throwing in the towel completely. In any given week, eggs are good/not good; ditto salmon/Omega 3s in fish and fish oils. Avoid fat/eat avocados, nuts, seeds. Vegan is good/bad; eat more fruit/avoid fructose; low-fat/high-carb vs. paleo vs. whole grains! Even kale is now being warned against! I’m considering just taking up an “air” diet, though I suspect a study would come along finding fault with that, too…

  30. Great article, especially when you write: “…that’s because readers don’t seek out health stories to satisfy abstract intellectual curiosities. They want to glean some kind of practical knowledge.”
    I’ve come to think that there are two audiences for science stories: people who are interested in the science, and people who are interested in practical knowledge that affects their lives. Reporters who write for a general audience write about the latter; science journalists, writing for an audience of like-minded people, write about the former.
    What’s missing is a way to bridge the two. That bridge, it seems to me, is explaining how science itself works. Over and over, until we, the public, understand it better.

  31. The “batch” idea would be good, if possible. It would be like a review article, but I suspect it’s rarely possible, given deadlines, etc.

    I think if the author of the journal article can be asked for the context, you’ll get something to quote that provides similar caveats, as long as 1) the study isn’t supported by an industry of some type; and 2) the author’s institutional press agent isn’t hovering about.

  32. A good science writer digs deeper into the research and on this issue many did not. The bottom line is “dietary resveratrol. ” which is something different that has been used on hundreds of studies involving large amounts of resveratrol based on David Sinclair`s work. Are you going to gain anything beneficial from dietary resveratrol.? No However, Sinclair`s previous work on Resveratrol was verified and substantiated last year. http://www.sciencedaily.com/releases/2013/03/130307145259.htm
    Secondly, there are hundreds of studies demonstrating high dosage resveratrol works. Sinclair just won the Time Award of the most 100 people who have influenced the world based on NAD+ and resveratrol research. NAD+ levels are restored through high dose resveratrol as well as his new research in using the compound separately. http://hms.harvard.edu/news/genetics/new-reversible-cause-aging-12-19-13
    I think that the response of Sinclair to the study in the LA Times yesterday say it all. ” “The levels of Resveratrol in the diet are negligible compared to the levels shown to work in mice and humans,” said Harvard University researcher David Sinclair.

  33. Reporting on single experiments is nothing more than marketing, unless the entire body of knowledge and theory that the experiments are embedded in is presented, along with the paradigm that the scientists in that community are operating under. I think a science wiki would be a great way to have all experiments presented to the public.

  34. I really have no problem with the journalist just reporting on the study as provided. Any subsequent information relative to the study is welcome but not necessary. Just point out to the reader that the science is still in its infancy and there’s more to come. It also helps to point out if the research has been duplicated by any other scientists or laboratories. I like being informed of areas where there is active research and don’t expect that each study is an end unto itself.

  35. A big study in the NEJM is news for the same reason that scientists want to get their studies published there. Journalism recognizes the article out of respect for the culture of science that has coalesced behind the idea that NEJM publication is a landmark of significance–even while both recognize that that level of significance is far from definitiveness. At the same time, scientists, in a way, depend on the journalist’s interest in their work. If science journalists stopped reporting what appeared in NEJM, arguably the journal would lose its significance to scientists. In short I agree with Virginia’s argument, but I also think that it’s pissing in the wind.

    People who lack experience with how science works or interest in science as an endeavor will find reasons to doubt scientific consensus if it interferes with the memes of their own thought-communities. The not infrequent contradictions or switches in scientific consensus offer them one opportunity to do this, but they have plenty of other dodges to embrace.

  36. see a few posts above by Allan Finegan
    I concur in this well written opinion.
    As an Doctor and Attorney with a masters degree in clinical nutrition, Mr Finegans post is entirely consistent with my own research and observations over the last 30 years in this field. Kudos!

  37. Thank you for your post. This is a complicated matter, where capitalism, culture, ethics, and the production of knowledge come together.
    What we need is critical journalism — like your own post. This applies to health “news” as well as to current affairs and sports. Journalists need to ask tough questions and help readers become critical learners.

  38. information is a two way street – how it is offered and how it is received. As a pediatrician, I know that no matter how good a job I think am doing presenting information on the safety and efficacy of vaccines, someone who needs to believe there is an easy answer for autism will recieve that information as further proof of how doctors just don’t get it, and someone who believes that “more is better’ when it comes to medicine will not even ask me what vaccine their child is getting that day – just go ahead and give it doctor. I could give you a hundered more examples. Yes we have to provided context as well as we possibly can, but we have little ability to control our patients’/readers’ ability or willingness to think critically. In my experience, scientific writing will only ever be as good as the critical thinking abilities of the reader.

  39. Thank you for this blog, Virginia. I really enjoyed reading it. It rings so true for me and my friends. Good nutrition/medical research is hard. Your readers are not equal. They are likely to have very differing expectation or motivation for reading you. I for one find it crucial to see a summary of the study method when reading about a new finding. Without it I could never know what to think of what is being reported. Unfortunately, often times details are left out. Contributing to appearance of contradiction where there is none. For example, I do not believe any nutrition research result based on food diary reported by the client, in the absence of dietitian (most are done this way). So, one requirement is that the clinical study must be done in-house, where the study participants are under continuous observation. Figure out what the most stringent criteria are and then measure the reports against them. At least then, you readers will appreciate the progress but at the same time be informed why this study might not be the last word. Look forward to reading you some more.

  40. I covered the Dipak Das fraud on his Resveratrol papers for a German Life Science Magazin, and it was the research for this article, that made me think exactly what you wrote here. So good to hear it from someone else. I have been thinking about this problem ever since and came up with timelines. I wrote down some of my thoughts here, if you are interested: http://faktoid.net/?page_id=18&paged=3
    and even tried to make an example (keep mouse on the text in the bubbles, to open the comment).
    Its in German though. Feel free to share your thoughts.

Leave a Reply

Your email address will not be published. Required fields are marked *