Supplements: Something Smells Fishy


In 2010, I wrote an article for ScienceNow titled “Fish Oil Fights Inflammation.” The article focused on new research showing that the omega-3 fatty acids in fish oil block inflammation in cells and fight diabetes in mice. Jerrold Olefsky, who led the research, even had an explanation for how they do this. This whole process hinged on a receptor found on immune and fat cells.

Olefsky and I spent maybe 30 minutes on the phone. By the end of the conversation, he had me hook, line, and sinker. Yes, his results were preliminary, but they were also striking. I went straight to the pharmacy and bought an enormous bottle of fish oil pills. I’ve been taking them ever since.

But recently I began to question my decision. Fish oil may help keep mice healthy, but the health benefits of fish oil supplements in humans are not so clear-cut.

Here’s a headline from 2008: Fish oil supplements help heart disease patients. Here’s a headline from 2012: Fish oil may not help heart disease. (Recent studies also suggest that fish oil pills don’t curb depression in heart patients or stave off mental decline.) Yet Medline Plus, a web site run by the National Institutes of Health, still says that fish oil is “likely effective” for heart disease.

Confused yet? Yeah, me too.

In addition to fish oil, I take calcium because my doctor told me to. The chewy, chocolate squares give me 500 milligrams of calcium and, as a bonus, 500 international units of vitamin D. In May, however, the U.S. Preventive Services Task Force issued a draft recommendation saying that younger women like me don’t need calcium or vitamin D supplements.

One can’t help but wonder, do we really need any of this crap? If a science writer can’t figure out what’s good for her and what’s bad, what hope does a person with no reporting skills have?

We have a serious communications problem. How can health reporters make sure that readers understand all the caveats? First, we can talk about the study design. Some studies are simply better than others. A clinical trial should carry more weight than an observational study. Second, let’s provide some context. What other studies have been conducted? How does this new study fit into the larger body of research? Too often decades of previous research are dismissed with a single line that refers to prior “mixed results.” Let’s dig in a bit more.

Part of the problem is that the science for many of these supplements really is uncertain. Strong hard evidence for a miraculous pill is hard to find. Gary Schwitzer wrote a fantastic guide to covering medical research. He begins it with a quote from Dan Greenberg:

The press, on its own, if it chooses, can make the transition from cheerleaders of science to independent observers… The journalistic trumpeting of medical cures on the basis of wisps of evidence, even though accompanied by sober cautions against optimism, deserves to be severely throttled back, in recognition of an unfortunate reality: though news is sold around the clock, major advances in medicine come along infrequently.





Share Button

12 thoughts on “Supplements: Something Smells Fishy

  1. I guess the most fool-proof way of reporting on the various strong, weak, and conflicting studies on a given question is to get an advanced grasp of meta-analysis and crunch the data on them all, taking into account the sample sizes of each, etc. Until then, it’s no wonder trying to juggle them all qualitatively feels unwieldy — it’s not the kind of calculation that lends itself to estimates.

  2. But Jessa, I never know how to trust those meta-analyses either. Their sample sizes are big but they’re putting together studies with different designs, definitions, purposes, what have you. As you know. So how do you personally do it?

  3. There are no easy answers. For years I took Vitamin E, then they said to stop, so I did. My pediatrician said to give my kids flouride, but it seemed like a lot, so I gave them their pills every 2 or 3 days. Years later I admitted this to her and she said, “That’s fine.” They have no cavities. Did I do the right thing? I really have no clue.

    I do think that research that could be of use in the pharmaceutical industry should be viewed with a more critical eye than research that is of no immediate monetary value to anyone. I wrote a fish oil/inflammation story in 1985 (Science News). Basically, it said to each oily fish and other sources of omega 3 fatty acids. Fish oil supplements went big a long time later. But your guy’s fish oil receptor shouts pharmaceutical application to me. That possibility is going to influence his level of enthusiasm and willingness to volunteer the counterarguments to his conclusions.

  4. Often, the conflicting info in studies has arisen due to the complexity of the plant’s properties (duh?). As you note it could be the role of journalists to question closely these details and avoid those bright headlines “If Steve Jobs Had Eaten This Plant, His Cancer Might Not Have Killed Him”. It should be handled by the researchers without such questioning but their funding often depends on having results that leads to additional funding (and validation of a commercially-valuable property).

    As a first cut on any research, I look at who paid for the research and then at the primary researchers and authors, where do they work and is that some academic sounding subset of the funder who also manufacturers the product? If a product is only researched by a few people who also publish a lot of articles, I check to see if articles are different studies or are the same study reported from different angles that gives the impression of strong support. The lack of a variety of different researchers is suspicious.

    But after analyzing the study funding, authors and specifics of the study design, methods, and data analysis, it is important to consider what was being evaluated and if that is all that matters about the end results: is this useful medically or for supporting health.

    Sometimes the chemicals (phytochemicals) are only present during certain growth periods or in certain plant strains or particular plant parts. Substrates plants grow in can alter properties. And availability in the body in vivo compared to in vitro skews the results even more.

    My favorite mixed message of the moment is Vitamin E that confounds research by showing great effects on heart protection in some studies and none in others. What researchers are now finding is that the form of Vitamin E might be key. The abundant and thus cheaper to provide form, alpha, seems to be the one that doesn’t help while the less abundant gamma does hold the protective abilities found in some studies. To further confuse this subject, it also seems that presence of both alpha and gamma forms diminishes the potency of the gamma form.

    Another current supplement dysfunction is curcurmin. Tons of research on this extract from tumeric root is creating a large amount of popular articles and supplements promoting curcurmin. But research from over a decade ago points out the problem of absorption and uptake in the human body due to curcurmin’s lack of water solubility.

    Formulas are now created that include chemicals that aid in curcurmin’s availability or that micronize it. I don’t know if the micronization really does make curcurmin more available as that term is such a buzzword. But the bottom line on curcurmin is that yes all the research on its medical powers seems correct; but in many of the popular products only 1% of what is in the product is actually metabolized by the human who buys and ingests it.

    I think for healthy people who live a healthy lifestyle, they can be a bit careless without damaging themselves. People with compromises, such as chronic illness, age, babies, etc might be helped by wasting a bit of money on unnecessary products to ensure that they do meet their needs.

    Now. How does one be sure that the perhaps unnecessary product doesn’t do harm? All the warnings about toxic metals in fish and radioactive tuna presumably from Fukushima radiation leaks don’t lead me to having interest in fish oil. Should I trust this molecular distillation process to really remove all the toxins? And with TCM products, contamination or mislabeling has caused trouble, even deaths.

    Sadly, in the US the agency whose role it is to require the research, evaluate it and determine what is and what is not safe and necessary for our health is the FDA. You know, the agency that allows drug companies to make false claims about what causes and relieves depression and to create advertisements for TV that spread the misinformation.

    So if you wish to really get to the heart of the problem, blame the FDA for not doing their job.

  5. As a consumer of science reporting can I ask that you include a summary of just what’s being announced. What did the study test and how? On how many people? With what controls and variables? Over how long? And of course, who paid for it?

    Just give us the facts and let us decide. If all we read is the headline, then we get what we deserve.

    And regardless of whether you think it’s dumbed down too much or too little, give us a link to the study. You would not hesitate to attribute a quote to a person, why not tell us the source of other sorts of information?

  6. I’m in the midst of reading Michael Pollan’s “In Defense of Food”, which gives the soundest nutritional advice I’ve ever heard: “Eat food. Not too much. Mostly plants.”

    Humans have survived for thousands of years without supplements, and died early because of infectious diseases. We are living longer on average compared to 1900, but this is a statistical fallacy — the life expectancy of a 65-year-old has only increased 6 years between 1900 and 2008. The reason we see life expectancy statistics surge over the course of the 20th century is because more of us are surviving infancy.

    I briefly considered supplements (multivitamins, calcium, protein, fish oil, omega-3, vitamin D, vitamin B12, etc.) but the research is too unstable. For every study that claims one thing, there’s another claiming the contrary. It seems easiest, and wisest, to let our natural tendencies guide us, and not rely on synthesized nutrients and manufactured “health food.”

  7. A big part of the problem is that we, as yet, understand very little about the billions of cascades of molecular mechanisms that are occurring in our bodies every instant – we just don’t know how most of it works yet.

    We’ve identified thousands of little pieces of biological mechanisms, and have developed drugs based on some of those – but do we understand the entire set of mechanisms affected by any one complex molecule? No.

    We’re still developing the tools to trace those interacting cascades of molecular mechanisms – and we’re quite a distance away from success there. Even if we did have the tools, we don’t yet have the computing power to model the incredible complexity we know is happening. That too is under development.

    So we rely on very crude statistical studies, where we give x number of people substance z and y number of people placebos, in double blind studies, and then check how many in each group get disease A, how many in each group get symptom B. Crude as it is, it’s the best we can do at this point, with the resources we have.

    Until we have the tools, funding and time to unravel how humans function at the molecular level, we have to put up with what we’re able to do. And we have to put up with a state of biological ignorance that is fairly profound, in spite of all that we’ve learned so far.

    We’re many decades away from understanding how any complex molecule interacts with all of a specific human’s biology. Get used to frustration!

Comments are closed.

Categorized in: Cassandra, Food/Drink, Health/Medicine

Tags: , , , , ,