More Science Talk
Welcome to Science Talk, the weekly podcast of Scientific American for the seven days starting February 13th, 2008. I'm Steve Mirsky. This week: a chat with epidemiologist Paul Marantz about how studies on lots of people's lifestyles lead to health recommendations and medical interventions. We'll have a Valentine's Day poem from a listener. Plus, we'll test your knowledge about some recent science in the news. Paul Marantz is professor of Clinical Epidemiology and Population Health at the Albert Einstein College of Medicine in New York City. He and colleagues recently published a controversial paper in the American Journal of Preventive Medicine. On Monday morning, February 11th, we talked in his Einstein office about the article and about the philosophy behind recommendations based on studies.
Steve: Dr. Marantz, good to talk to you today.
Marantz: Hi, Steve.
Steve: Very interesting paper you have here, "A Call for Higher Standards of Evidence for Dietary Guidelines." Let's cut right to the quick and then we'll back up a bit; but your basic thesis is?
Marantz: Our basic thesis is that the standards that had been applied to determining and promulgating dietary guidelines for all Americans have been insufficient to protect against the possibility of harm and in fact our analysis suggest that there indeed may be harm that can be an outcome of these guidelines. And once that's considered, the issue of standards of evidence becomes much more pressing.
Steve: And specifically, you're talking about the dietary guidelines about fat, as an example.
Marantz: That's the example we focus on. I think, the general concept of dietary guidelines is one that we explore, because we were curious as to why we then are in this business? Or why [is] the government is in this business? So by analyzing and focusing on the dietary fat guidelines, which we are not the first to do, others Gary Taubes and others have made that point in the lay press, and it has been in their professional literature as well, but we provided another data analysis consistent with the notion that focusing on fat led Americans to eat more calories overall, [and] has contributed to our obesity epidemic. And in light of that evidence, we really should be extremely cautious and careful when issuing guidelines.
Steve: Right. So the specific point that's controversial is that the dietary guidelines that were put into effect by the government in an effort to get people to cut down on their fat intake actually contributed to the obesity epidemic, and you point out this [out], you know, its very simple math—people did cut down on their fat calories as a percentage of total calories by increasing their total calories.
Marantz: That's it. Yeah, you can change the percentage or the proportion in two ways. You can reduce the numerator—how much fat people eat in total—which was clearly the goal of these guidelines; or you can also get the same effect proportionately by increasing the denominator—how much total calories you eat, I mean the total calories you eat.
Steve: Right. So now you have people [saying], "Well, I have cut my fat percentage down to below the 30 percent that the government recommended," but they are eating 3,000 calories a day, instead of 2,500.
Marantz: We can all remember, and in fact we still do, wolf down these low fat snacks that seem to us to be healthful or safe because we were really all taught that if the food was low fat it was healthful—that was the inference that we were led to make. By the way, this is not ancient history, the first official published dietary guidelines for all Americans were published in 1980, and it is now a legal mandate that they have to be reissued every five years. You know, reevaluate the evidence as it's done and reassure the guidelines. Then the focus was on cardiovascular disease, so a single-minded focus or at least a primary focus on dietary fat made sense with a cholesterol hypothesis. And of course we weren't focusing on total calories, we were focusing on dietary fat. The irony is, now that we have the obesity epidemic, suddenly calories have become the issue.
Steve: Go back a little bit to the idea of dietary guidelines. You point out in the paper, [the] government first issued nutritional guidelines in 1894, but these dietary guidelines as you say are kind of new. The original guidelines—well why don't you explain the difference?
Marantz: Well, they made great sense in the turn of the century when public health professionals were trying to meet the mission of public health, which is defined as assuring conditions in which the public can be healthy—that's how public health is defined. And in a time when malnutrition was a problem and, the emergence of nutritional science was helping public health professionals understand what needed to be in the diet in order for people to be healthy, to avoid deficiencies. It made good sense to issue information about the way in which vitamin C can prevent scurvy and that sort of stuff and that's where I guess, the minimum daily requirements came from. Over the 20th century, we did see our nutritional concerns move from issues of deficiency to issues of excess; and in that shift we saw a focus instead on the sorts of problems that excess leads: to coronary artery disease, diabetes, obesity, those sorts of concerns.
Steve: In your paper, you discuss something that doesn't get really talked about too much among lay people and that's one of the key, kind of, philosophical foundations of epidemiological studies and policy recommendations and that's this idea of small changes in risk for individuals winding up making large changes in outcomes in populations. That's something that really informs a lot of the decision making. So let's talk about, you know, what we see it in a cutback on your salt if you have hypertension and in the dietary guidelines, too, with cholesterol levels and diet. So let's talk about that a little bit, this idea of little changes for people making big changes in populations. One small step for a man, one giant leap for mankind!
Marantz: There you go! I'm glad you picked up on that because I think that really is the crux of this argument, and I don't think it is well appreciated. We can date or at least credit the important insight to a great British epidemiologist, Sir. Jeffrey Rose, who put forth what he called as a population approach to prevention, a different sort of strategy of preventive medicine. And what he pointed out—and this was I guess during the, I think his classic papers were in the '70s or '80s, you could find that out, I don't remember that off the top of my head—where he pointed out the great successes we've had in preventive medicine through our traditional high-risk approach. Hypertension is diagnosed; you get your blood pressure checked; we decide based on lots of good evidence that if your blood pressure exceeds a certain value, it deserves to be treated; we may have different kinds of treatment protocols, but we identify you as high risk because your blood pressure is high. We treat the blood pressure and in so doing we make the inference that we have lowered your risk and there is good data behind that.
Steve: There is now, when it was first done, is it not true that when these kinds of widespread recommendations and treatments went into effect, we didn't know that lowering blood pressure would decrease your risk of stroke or other cardiovascular issues? We just assumed it. Now we have data that show that that's the case, but the recommendations and policies went into effect, based on the correlations before we had data showing that there was causation.
Marantz: Right. And in fairness Steve, we can only use the data we have, and this is in fact, I think, a great triumph of scientific enquiry that led to a very logical sequence of events. There was good physiologic and observational data to support the belief that higher blood pressure would lead to strokes and heart attacks. That was then followed up by epidemiological observations that showed that association. That creates a hypothesis that says, okay if high blood pressure leads to more heart attacks, lowering blood pressure will reduce heart attacks, but you're quite right, that doesn't prove it, but now you have blood pressure as an important target for intervention. Drugs were then developed and approved where the only thing initially they could show is that they could reduce blood pressure. Okay that's fine, but that's only an intermediate endpoint. The next step ...
Steve: Right, because the important thing is we want to stop disease.
Steve: Not just the markers for disease.
Marantz: Right, and hypertension is a funny case, because we all think of it as if it were a disease, but it's really not. It's an arbitrary cut point on a physiologic variable—your blood pressure—and above a certain cut point, we say you've got this thing that we call hypertension; below it, you don't, but we need that, we need to operationalize this clinical behavior, this high-risk approach that I'm referring to, and then ultimately the high quality studies, randomized clinical trials, looking at the important end points, heart disease, and stroke and mortality were done and did demonstrate that, certainly with certain classes of drugs, they are treating high blood pressure and reduced those bad outcomes. So now we've come full circle and that belief has been proven and we continue to treat high blood pressure and that's an effective approach. But what Rose pointed out is that high-risk strategy by identifying say the top 25 percent of the population in terms of blood pressure and risk still misses 75 percent of the population who are also at risk. There is no free lunch in this and[or] whatever the right phrase is.
Marantz: No one's immune.
Steve: There is no free, salt-free lunch.
Marantz: No free, salt-free lunch. Nobody is immune from cardiovascular disease, so we all are at some risk.
Steve: Right. Let's spell it out a little more clearly. Most people who have high blood pressure or what's considered to be high cholesterol will not have a cardiovascular disease.
Steve: And many people who have normal blood pressure or normal cholesterol levels will.
Marantz: Thank you.
Steve: So that's really what's going on here, but the population versus individual conflict is really what's addressed in the studies and Rose really addressed it. So now talk about what he wanted us to do.
Marantz: Exactly. You imagine this bell shape curve, and at the upper end of the distribution—I don't know if this is helpful; it's helpful to me.
Steve: Well, I think the listeners can see the bell-shaped curve in their mind's eye; and we're talking about over at the right where it gets skinny.
Marantz: Exactly, so up at that higher end, you've got folks who are at greater risks, more likely to have these bad things happen to them, but there are a few of them because, as you point out, it gets skinny over there. And what about that big group in the middle? So what Rose said is, "Well, it's great to take that upper tail and shift it back to the left, but what if we could shift the entire curve to the left, ever so slightly? It would actually have a greater effect on the population on overall health"; and this was really a powerful insight. So a population based approach, even though the actual incremental benefit at the level of the individual is minute and arguably it's unmeasurable. In fact Rose referred to this concept of the prevention paradox, which is that very powerful preventive interventions will have no measurable effects at the individual level, but big effect at the population level. So, in fact he talked about salt, you mentioned that. He said, "Imagine, if we could reduce the amount of salt people eat, which would shift the blood pressure curve over to the left just a bit, that would have a powerful impact magnified over the entire population." He said in his paper, "This all presumes that the change is safe." So I want to be very clear, this concept that by eating less salt as a population we will reduce our blood pressure and therefore improve our risk of cardiovascular disease, that's a hypothesis that is [has] actually not been proved. My senior coauthor on this paper, Dr. Mickey Alderman, has actually written quite a few papers challenging that dogma and in fact pointing out that there are inherent risks involved in reducing salt intake, so that these things that are so ingrained in us as to feel like truisms often are not and certainly may not be [true], and I think what is important—and the reason we wrote this paper, but it is also something that I recognize as really challenging to understand—is that its much more confusing than we would like to believe.
Steve: Can you give us a number, just to make this point even more clearly? Do you have to treat something like 6,000 people with high blood pressure in order to get one person who does not have a cardiovascular event because of it?
Marantz: Yeah! You know, there is this epidemiologic concept called "number needed to treat".
Marantz: And it all depends on how long you're treating them, in which trials you are looking at. The number that I remember—and this is going back a while—was that you needed to treat 850 mild hypertensives for a year in order to prevent one stroke.
Marantz: But you know, and I guess as a sound bite that sounds, you know, "Gee, where's the efficiency there?" I mean we're putting so many people at risk of the treatment for so little benefit, but the fact is this is the way preventive medicine is always practiced. Outcomes that we are trying to prevent tend to be rare and take a long time to occur, and we never can be sure which of the many people we are trying to prevent it in are actually going to be the one who's going to get it.
Marantz: So, we have to use a shotgun approach right now and, you know, there is a lot of talk about the promise of genomic medicine, lets hope we get there; in fact, we'll get there but we're not there yet and we're still using the clinical shotgun approach, we're also using a clinical shotgun, a public health approach and that's what we're really trying to tackle in this paper.
Steve: Right! You quote Hippocrates, "first do no harm", and you wonder whether or not the issuance of these guidelines actually does more harm than good.
Marantz: Well, and even once you've raised that question—which is what we're trying to do—I think that does change the discussion. It is easy to go ahead and tell people what they ought to do when we believe in our hearts that there is no downside risk to that and I think we have believed it; I've believed it. So, this analysis and this paper came from actually several years of debate between me and my senior coauthor, Dr. Alderman, until we realized what it is that we were arguing about. In fact it started from a time where I said in a lecture that we were giving together that "the standard of evidence for public health guidelines is lower than that for clinical recommendations," that was a statement I made. Mickey pulled me aside after the lecture and said, you know, "I completely disagree with you" and we ended up in a back and forth on this and it took a while before we realized that I was describing the standards that are applied, [and] he was talking about the standards that ought to apply. And once we realized that that was the cause of our debate we began the process of writing this paper.
Steve: And in some ways this paper is designed to start a conversation.
Marantz: That is our goal. We recognize that it is challenging and may be provocative. It's not meant to be obnoxious but it's saying in a way that is challenging—because there aren't easy answers that come from this—"okay, so and then what should we do?" And I recognize that. I mean the best we could come up with than what we see in the paper is rather than focusing on the guidelines, the sound bites, the instructions, just provide the information in all its complexity and then allow people to make their choices.
Steve: Or provide no guidelines at all, that's the more controversial?
Marantz: Well, I mean that's kind of what I'm saying there. Not providing guidelines doesn't mean we don't provide information.
Marantz: And I do understand and share the concern that, you know, without some sort of governmental academic expert [vetting] wedding people are going to hear all sorts of stuff and there are a lot of people with access [axes] to grind or pills to sell and I think it is appropriate and there at least be someway to get information that you can trust is reliable. That's a worthy go, but unfortunately the reliable information often leads to, "Well, this is our best guess at the moment, but this may change or the evidence is okay on this one but really not strong"; that level of hedging and caveats is un[a]esthetic, but we think you need to have it.
Steve: And it's something that we are comfortable with within the scientific community, within journal articles, but becomes something that people are less comfortable with when they are giving out these governmental summaries of the findings and what to do with that, so things gets stripped down and people just say "Make sure that less than 30 percent of your calories are fat" rather than discuss details.
Marantz: Right, well and stripping it down is, a beautiful example of that in the case we're discussing is the food pyramid. What better, nice, simple way to distill all these reams of complex data to a point where you can say, okay look at this image and what we have is this base of a pyramid which is filled with bread and rice and pasta and cereal and all that stuff which we are being told is okay because that's the base of a healthy diet and then at the top is the oils that you have to use sparingly, oils and fats, and that really I think was the take-home message of that very effective mass marketing campaign. I know my kids were brought up on it and I got to tell you I had some very obnoxious little kids around the dinner table telling me what they had just learned that day in health class and it really used to annoy the hell out of me. I got to tell you.
Steve: (laughs) So you have a whole nation that's carbo-loading all at the time.
Marantz: Well, I don't know what they're doing now, because they are getting confused.
Steve: That was the old pyramid.
Marantz: Yes, that's right.
Marantz: Well, the new pyramid, have you looked at that?
Steve: Yeah, I have.
Marantz: Yeah, and I think that's wonderful because it's completely incomprehensive. (laughs) If you look at that thing you have no idea what it actually wants to say. So to my mind that's a great step forward. The only really clear message is that you should walk up the pyramid. Right. There is the guy climbing it.
Marantz: So, you know, they're promoting exercise that's a clear message. The rest of it, I don't know what they're saying.
Steve: Right. The pyramid is in the September issue of Scientific American by the way, if anybody wants to look at it. You did start a conversation with this paper because the same issue of the journal also carries a response by Marion Nestle, who we have had on the podcast and [Paul Raeburn] and they take you to task for some things and then you take them to task in your response. Everybody is very civil to each other as they rip out each other’s livers.
Marantz: Yeah, well and I do think it comes from a, you know, a really collegial disagreement and, you know, this is why I love academic medicine; and we do get to disagree with each other, but we tried to do on the strength of our arguments not on vitriol.
Steve: Dr. Marantz. Thanks very much. I appreciate your time.
Marantz: Thank you Steve.
Steve: Marantz'[s] article, the response and the author's retort are all available online free; the whole package is actually very readable and entertaining. Just go to www.snipurl.com/paul-SciAm for the article. The responses are at www.snipurl.com/paul2-SciAm and www.snipurl.com/paul3-SciAm; you guessed it.
Listen to this ... (whistling sound) Amazingly, that sound came out of a human being—Brazilian singer Georgia Brown. SciAm magazine recently ran an article by Ingo Titze on how the human voice produces all the singing sounds it does; it's free at www.snipurl.com/sing-SciAm. Take it home, Looch. (Georgia Brown's whistling song)
Now it is time to play TOTALL……. Y BOGUS. Here are four science stories; only three are true. See if you know which story is TOTALL……. Y BOGUS.
Story number 1: Most of the money spent trying to alleviate back pain appears to be wasted.
Story number 2: Lots of kissing on Valentine's Day. Oxytocin is a hormone involved in feelings of bonding and when a man and woman kiss, their oxytocin levels naturally rise.
Story number 3: Paleontologists have found the fossil of a teeny-tiny pterodactyl.
And story number 4: The sun-like star Tau Bootis flipped its magnetic field from north to south sometime during the last year.
Time is up.
Story number 1 is true. It looks like most money spent treating back pain is wasted. Research by Richard Deyo and colleagues at the University of Washington, published in the Journal of the American Medical Association finds that spending on back pain has gone up 65 percent since 1997 but without much effect. You can find the August 1998 Scientific American article "Low Back Pain" by Deyo—that's D-E-Y-O—at www.SciAmdigitial.com.
Story number 4 is true. Tau Bootis did flip its magnetic field from north to south in the last year. The finding appeared in the Monthly Notices of the Royal Astronomical Society. Actually our sun's magnetic field changes its direction every 11 years, but this is the first time that such a change has been observed in another star.
Story number 3 is true. 120 million years ago lived a pterodactyl the size of a sparrow. It was toothless and had curved toes the better for perching in trees. The find was announced in the online edition of the Proceedings of the National Academy of Sciences [USA].
All of which means that story number 2, about oxytocin levels rising in both partners during a male-female kiss, is TOTALL……. Y BOGUS. Because a study found that oxytocin levels rose in the males, but actually fell in the females. The researchers think that women require more than kissing to feel emotionally connected or sexually excited; for example, the experimental set up may not have provided a romantic atmosphere. Of course, that didn't stop the men. For more on kissing, you can read the article "Affairs of the Lips: Why We Kiss" at www.SciAmmind.com.
Speaking of Valentine's Day, I was a guest on the Skeptic's Guide Podcast a while back and I mentioned a letter from a reader to the magazine and the letter began "I am not scientifically smart." So, you know, we talked about how seriously we would take the rest of such a letter. Anyway, one of the listeners to that podcast and to this one, Christine Jones, out in San Francisco was inspired by the opening of that letter and wrote a poem for Valentine's Day. She sent it in; I liked it and I'm going to read it to you now. It's called "Argument from Ignorance'".
(Poem by Christine Jones, Read by Steve Mirsky)
I'm not scientifically smart,
Or keen enough to break apart
The covalent bond that binds my heart to yours,
And leaves me thrilled with a sense of strange adventure.
I am not scientifically wise
To decipher the message in your eyes,
Or thus employ my own two spies,
To cox your vitreous body into infinite regression.
I am not scientifically sage,
And despite the advancement of my age,
I would fling myself down at your bipedal feet
And dissolve in hot cesium vapor.
I am not scientifically shrewd,
But your sweet voice puts me in the mood
And surrounds my throbbing superior vena cava with a modulation of its exquisite softness.
I am not scientifically bright,
But with some persuasion I think I might
Lay photons on a sensitized silver hyalite plate,
And stamp myself upon your cerebral cortex.
I am not scientifically clever;
To possess you in totality will happen never.
And I will remain doomed forever.
To orbit your heart like a cast away.
Little unrequited love on Valentine's Day, but that happens, too.
Well, that's it for this edition of the weekly SciAm podcast. You can write to us at podcast@SciAm.com and check out numerous features at www.SciAm.com including the latest science news, articles from SciAm, SciAm Mind, and SciAm Body , not to mention all our podcasts. For Science Talk, the weekly podcast of Scientific American, I'm Steve Mirsky. Thanks for clicking on us.