Here we go again. No doubt you’ve seen on T V or read in the news that meat causes colon cancer. At least that’s the take home message you got if you saw any of these ‘news’ reports or read any of the articles about a new study published in the Journal of the American Medical Association (JAMA) this past week. Let’s take a look at what is really going on.
The study was pretty simple. Researchers looked at the diets of over 1000 subjects who had stage III colon cancer and who were enrolled in a chemotherapy trial. These subjects filled out food frequency questionnaires (FFQ) during their chemotherapy then again six months after. Researchers then stratified the data from the FFQs into what they called a Western dietary pattern and a prudent dietary pattern. After a little over 5 years of follow up about a quarter of these subjects had died from their colon cancer. Significantly more of the patients following the Western dietary pattern died than did those who followed the prudent dietary pattern. Therefore, say these researchers, those who have advanced colon cancer should avoid a Western dietary pattern in favor of a prudent dietary pattern in order to reduce their chances of dying from their disease. And the implication is that all of us should follow the prudent diet rather than a Western diet to maybe avoid getting colon cancer at all. And, as you shall see, what these researchers really want us to avoid is red meat.
Nice and tidy. Problem is that it’s all BS. And it’s the kind of BS that infuriates me because of the dishonesty involved. Let me show you what I mean.
The statistics involved in this paper are convoluted beyond belief. Very few people are going to go through the charts and tables and analyze these statistics as presented. I’m a pretty fair hand at statistics, and I didn’t want to spend the time to laboriously go through these to see if they were correct. I didn’t spend the time because it doesn’t matter if they are correct or not, so why bother. And that’s not to even mention that the nutritional data making up these statistics was derived from FFQs, which are the most notoriously inaccurate way to get nutritional data imaginable. The entire study is an observational study, and as such is absolutely worthless in determining whether the Western diet or the prudent diet or any other kind of diet causes colon cancer or anything else. But before we get into what an observational study really is and why it’s worthless, let’s look at a couple of other factors.
First, the researchers themselves cleverly defined the Western dietary pattern and the prudent dietary pattern so they could determine what made up each one. The prudent dietary pattern was filled with fruits, leafy vegetables, cruciferous vegetables, tomatoes, fish, poultry, wine, tea, and legumes. The Western dietary pattern was dairy, refined carbohydrates, red meat, sweets, desserts, margarine, processed meat, potatoes, French fries, snacks, beer, liquor, eggs, and sugar beverages among others.
Now this divvying up of all these foods into Western and prudent dietary patterns would be okay with me if that’s the way they reported it. I mean all the crap they put in the Western dietary pattern category is pretty much what makes up the typical American diet, and I, of all people, certainly wouldn’t be surprised to find that people following such a diet had their cancers worsen, but that’s not what the researchers reported. They focused on the red meat part of the Western diet in all their reports. It was made to sound like the red meat in the Western dietary pattern was the cause of all the woes, not the sugar, potatoes, sweets, high-fructose corn syrup sweetened beverages, margarine, and all the rest of the crap. Just the red meat. But before we get into examples of that, let’s look at what an observational study is and why these studies are worthless for determining the cause of anything.
Let’s say I come up with the idea that eating ice cream causes multiple sclerosis (MS). To study this I recruit 10,000 subjects and follow them for 5 years. Over this 5 year period I take nutritional surveys by having these subjects fill out detailed food diaries for 3 days every three months, and I have trained interviewers go over these food diaries with the subjects to make sure they included everything. I gather all this data for 5 years, then I look at the amounts of ice cream eaten by all the subjects, and I divide the subjects into quintiles (fifths) depending upon how much ice cream they ate. The top quintile would be the 2000 subjects who ate the most ice cream; the bottom quintile would be the 2ooo who ate the least; and the other three quintiles of 2000 subjects each would fall in the middle.
Now that I’ve got this data, I look to see how many of the subjects in each quintile developed MS during the course of the study. Let’s assume that 100 subjects in the top quintile of ice cream consumption developed MS and only 30 people in the bottom quintile developed MS. The number of cases of MS from the middle quintiles were between 90 and 4o with the higher quintiles developing MS at greater rates than the lower ones.
I could then put out a press release proclaiming that ice cream causes MS, which would be picked up by the press, and, before you know it, I’m on CNN and all the networks talking about my study. Life is great. I’m on TV. But hold on. Whoa, there, Bozo. Let’s take a few steps back and look a this study just a little closer.
The study I did was what is called an observational study. I didn’t give these subjects ice cream, placebo, medications, anything, I simply observed and tabulated their diets over a 5 year period. Those that ate the most ice cream had higher rates of development of MS, but the MS can’t be laid at the doorstep of ice cream consumption because there are too many other factors involved.
Maybe those who are prone to MS are, for whatever reason, driven to eat more ice cream. Maybe those who ate the most ice cream also ate the most cake, and we didn’t even look at cake. Maybe cake is the cause. Maybe those who ate the most ice cream also smoked the most. Maybe those who ate the most ice cream were the least active, and maybe lack of activity predisposes to MS. Maybe those who ate the most ice cream ate the fewest tomatoes, and maybe tomatoes are protective against MS. I could go on and on with the maybes coming up with countless scenarios showing that it isn’t really the ice cream that causes MS even though those who ate the most developed the most MS. The point is that in an observational study like this fictitious one I just came up with – which is no different than the JAMA study on the Western dietary pattern/prudent dietary pattern and death from colon cancer observational study – is meaningless in terms of what causes anything.
What observational studies are good for is creating hypotheses. For example, let’s say I did the above study and I was just looking to see which subjects developed what after 5 years. At the end of the study I discovered that a bunch of people had developed MS. I looked at all the different foods all the subjects ate and I noticed that the ones who ate the most ice cream seemed to develop MS at the greatest rate. So, I come up with a hypothesis that ice cream causes MS, then I get together the funds to test this hypothesis.
To determine if this hypothesis is valid I recruit 10,000 more subjects. This time I randomize them into two groups so that the ages, sexes, heights, weights, and any other thing I can think of are the same in both groups. Then I create a placebo that looks like ice cream, tastes like ice cream, but has no calories and isn’t really ice cream. (Actually, if I could come up with this placebo I would abandon the study, get filthy rich, and move to my own island.) I would then box the placebo and the real ice cream in identical cartons. I would have a staff of people who gave the ice cream or placebo to the subjects at regular intervals. The subjects would be instructed to never eat ice cream on their own, but to only eat the ice cream they were provided during the study.
I would then have a double-blind, placebo-controlled study. It’s double blind because the people administering the ‘ice cream’ don’t know whether its ice cream or placebo and the subjects receiving the ‘ice cream’ don’t know if it’s real ice cream or placebo.
After 5 years I could again determine how many subjects had developed MS and could compare it to the amount of real ice cream eaten by those subjects. If this time the results were the same, then I would have some reason to say that ice cream may indeed cause MS. But this would be a real, double-blind, placebo-controlled study, not an observational study.
Typically when hypotheses arise from data provided by observational studies, these hypotheses are destroyed when they are evaluated using a real study.
For the JAMA study to be able to say that the Western dietary pattern causes recurrence and death to a greater degree than does the prudent dietary pattern, the researchers would have had to randomize the patients into two groups, then give one group the Western diet and the other the prudent diet. And keep both groups on their respective diets for the 5 years, then see what happens.
And the authors of this study indeed know and understand that these observational studies don’t mean squat in terms of causality and admit as much. They write these words buried in the back of this study:
Because this was a observational study, causality cannot and should not be drawn from these data. [my italics]
So, the authors know and admit that no conclusions in terms of causality can or should be drawn from this study. But they go ahead and do just that with this statement that is the very next sentence after the one above:
Nonetheless, the data suggest that a diet characterized by higher intakes of red and processed meats, sweets and desserts, french fries, and refined grains increases the risk of cancer recurrence.
So, you fess up that the study doesn’t legitimately say squat, then you say in the very next sentence that, well, we don’t care about the truth, we think it really does matter. Because, after all, no one cares about a meaningless study. You can’t issue a press release about a meaningless, observational study. Can you? The authors and the folks at JAMA apparently think so.
Here is the press release from JAMA.
DIET HIGH IN MEAT, FAT AND REFINED GRAINS ASSOCIATED WITH INCREASED RISK FOR COLON CANCER RECURRENCE AND DEATH
Patients treated for colon cancer who had a diet high in meat, refined grains, fat and desserts had an increased risk of cancer recurrence and death compared with patients who had a diet high in fruits and vegetables, poultry and fish, according to a study in the August 15 issue of JAMA.
Previous research has indicated that diet and other lifestyle factors have a significant influence on the risk of developing colon cancer. However, few studies have assessed the influence of diet on colon cancer recurrence and survival, according to background information in the article.
Jeffrey A. Meyerhardt, M.D., M.P.H., of the Dana-Farber Cancer Institute, Boston, and colleagues examined the influence of two distinct dietary patterns on cancer recurrence and survival in a group of 1,009 stage III colon cancer patients (cancer present in the colon and lymph nodes) enrolled in a clinical trial of postoperative chemotherapy in addition to other treatment. Patients reported dietary intake using a food frequency questionnaire during and six months after supplemental chemotherapy. Two major dietary patterns were identified, prudent and Western. The prudent pattern was characterized by high intakes of fruits and vegetables, poultry, and fish; the Western pattern was characterized by high intakes of meat, fat, refined grains, and dessert.
Patients were followed up for cancer recurrence or death. During a median (midpoint) follow-up of 5.3 years, 324 patients had cancer recurrence, 223 patients died with cancer recurrence, and 28 died without documented cancer recurrence.
The researchers found that a higher intake of a Western dietary pattern after cancer diagnosis was associated with a significant increase in the risk of cancer recurrence or death. Compared with patients in the lowest Western dietary pattern quintile (bottom 20 percent), those in the highest quintile (top 20 percent) experienced a 3.3 times higher risk for cancer recurrence or death. Patients in the highest quintile of Western dietary pattern were 2.9 times more likely to have cancer recur than those in the lowest quintile. Similarly, a significantly higher overall risk of death with increasing Western dietary pattern was observed. In contrast, the prudent dietary pattern was not significantly associated with cancer recurrence or death.
“Studies have shown an improved disease-free survival among patients who receive adjuvant chemotherapy following the surgical resection of stage III colon cancer. This is the first study, to our knowledge, in a potentially cured population of colon cancer survivors to address the effect of diet. Because this was an observational study, causality cannot and should not be drawn from these data. Nonetheless, the data suggest that a diet characterized by higher intakes of red and processed meats, sweets and desserts, french fries, and refined grains increases the risk of cancer recurrence and decreases survival. Further analyses are under way to better delineate specific nutrients or food groupings that may have the strongest association,” the authors write.
Do you see anything in this press release that says no conclusions can or should be drawn from this study since it is an observational study? I didn’t see anything like that. And did you notice how it is no longer a Western dietary pattern, but is now a meat, fat and refined grains diet. And note what is listed first: meat. Most analyses of the Western diet show that the single largest contributor of calories is sugar, which I’m sure is the case in this Western dietary pattern, but the authors and the press release writers for JAMA single out and lead the press release with meat. Not sugar. Not high-fructose corn syrup. Meat.
What’s even worse than the press release is the video that goes along with it. In this video a survivor (so far) of stage III colon cancer, John, is shown cycling and going about his life. A voice over announcer talks about the diagnosis and treatment of John’s colon cancer. Then the announcer says
As you’d imagine, after that, John would do a lot to stay healthy. Turns out what he eats, or doesn’t eat, could really help.
This while John is shown putting fish on the grill. Then cut to Dr. Jeffrey Meyerhardt, the lead author of the study, who says:
It’s not really increasing the amount of fruits and vegetables but really trying to reduce the amount of red meat intake and fatty foods and sugary, ‘desserty’ foods, that seems to be protective for colon cancer recurrences and survival.
Note the emphasis on red meat and fatty foods. As Dr. Meyerhardt finishes his line the video cuts to a steak on a grill, then to someone cutting a steak. The announcer then says:
That higher fat diet Dr. Jefery Meyerhardt describes is called a Western pattern diet. He and colleagues at Dana-Farber Cancer Institute studied that pattern in colon cancer patients. Their findings appear in JAMA, Journal of the American Medical Association.
While the announcer is talking there is more footage of red meat on a grill.
There is some more blather, then the announcer says:
Stage III colon cancer patients who ate high amounts of a Western pattern diet were about three times more likely to have recurrent cancer, or to die, compared to patients who ate less of those types of foods.
More footage of John and his wife eating fish and vegetables. Then the announcer again:
The study says improved outcomes are more likely if stage III cancer patients eat the way John does…lots of fish, chicken, brown rice, and less Western pattern foods, like red meat.
Actually the study says no such thing. The study says that “causality cannot and should not be drawn from these data.”
John finally speaks. He is eating his ‘healthful’ non-Western pattern diet along with his wife, and he says:
To cut down from one steak a week to one steak a month is not a big deal.
As you watch the video you notice that right at first the overall Western pattern diet is described as being filled with red meat, fatty foods, sweets, and desserts. Then as the video rolls on the Western pattern diet transmogrifies into the red meat diet. Anyone watching this video would come away thinking that eating red meat will cause colon cancer to worsen. And that if anyone who has colon cancer eats meat, they’re going to die.
And all this from a study whose authors tell us that the data cannot and should not be used to determine causality. This entire episode is an education in how this nonsense spreads. And another reason to never, ever believe anything you see on television or read in the papers about medical studies. Most of it is pure hogwash.
Why, you may ask yourself, do these highly trained scientists buy into all this when it’s so patently false? Because they want to. Remember the post about the book Stumbling on Happiness? It takes very little evidence to persuade someone of something he/she already believes is true. All of these people believe red meat is bad for us despite there being absolutely no evidence showing such. So it doesn’t take much to prove to them that it really is bad. In this case, it takes no data at all. Merely the suggestion of data. Pitiful.
Here is a link to the JAMA video from which I quoted above. Warning. It takes forever to download this thing, which is why I quoted from it instead of merely linking to it. It took me about 5-6 minutes, so don’t give up if you really want to see it. It is a masterpiece of anti red meat propaganda.