Last Sunday’s Age carried two articles on red meat by Melinda Houston. The first concerned red meat as a health risk, the second concerned the environmental impacts of red meat and will have to wait for another day.
After a few paragraphs presenting evidence against red meat, Houston wheels out the defenders. And doesn’t she have some heavy hitters! CSIRO Professor Peter Clifton, CEO of the Cancer Council, Professor Ian Olver, Emeritus Professor of Nutrition, Stewart Truswell and topped off with a Meat and Livestock (MLA) nutritionist Veronique Droulez. Houston quotes some statistical arguments from her stars and then throws in a nice little scare story about a well known music personality who ended up with gall stones, allegedly from a bad vegetarian diet.
In her opening paragraphs, Houston refers to my article in Crikey last year in which I described eating red meat as like smoking through your an-s. Houston says I “declared lamb chops the nicotine of the 21st century”. But this totally misrepresents my comment as a metaphorical flourish. Rubbish. I never mentioned either lamb chops or nicotine because my statement was not a metaphor. It is quite literally true.
If you did manage to inhale cigarette smoke through your an-s (don’t try this at home), the chemicals in the smoke which would damage the DNA in your bowel, possibly leading to cancer, are the same kind and in similar concentrations as those which end up in your bowel after a red meat meal. The nice thing about writing on the web is that you can check my statements, so here you are, go look in the journal “Carcinogenesis” and check it.
Clifton and Olver run the tried and trusted tobacco industry approach of saying that no-one has proved causality. Olver says population studies linking red meat with cancer aren’t enough, but merely give “impetus for further, more specific, research”. Absolutely. Which is exactly what researchers around the world have been doing for at least 10 years. The World Cancer Research Fund (WCRF), mentioned by Houston, summarise the state of research every 10 years and in their 1997 report they had plenty of population data, quite a bit of mechanism data, and they thought that all meat probably caused bowel cancer. But now, after 10 years of more specific research and better population data, white meat (chicken and fish … not pork) has been exonerated and red meat has been found guilty. The “probably” is gone. If Olver hasn’t read this work, he gets out too much.
Those at the CSIRO who haven’t sold a million copies of a book touting the benefits of red meat know all about this work. They have developed a special resistant starch which can reduce the DNA damage associated with the red meat in the diet their organisation promotes. They have patents and everything. Why bother if the damage is harmless?
Clifton is famous with co-author Manny Noakes for spruiking the benefits of lean red meat in the CSIRO Total Wellbeing Diet (TWD). Clifton adds to Olver’s shawl of causal doubt with talk about “confounders”. What’s a confounder? Suppose meat eaters get lung cancer more frequently than vegetarians, does it prove that meat causes lung cancer? No. You need to allow for the possibility of different smoking rates. Smoking is a confounder. If you know it’s a confounder, then correcting for it is easy … for any statistical geek. It’s the unknown confounders that can always be a problem. Which is exactly why researchers also look at mechanisms. To infer causality you need proof at multiple levels, which is exactly what WCRF required and found.
In any event, and in my view, it’s pretty rich for Clifton to talk about confounders and the difficulties of proving causality. His TWD diet is famous for being a scientifically proven weight loss plan … it says this in really big print on the cover. What evidence did he need to have to prove his diet works? Was he worried about confounders in his work? We shall see.
While the WCRF cite population studies with hundreds of thousands of people, masses of lab work and some brilliant controlled clinical work in volunteers, Clifton used a controlled study of about 100 women over 12 weeks with a 12 month follow up period.
This was enough to convince the CSIRO of the efficacy of its diet. On page five of the record selling book, the authors claim the TWD is better than a conventional high-carb/low fat diet:
Once we had established that the CSIRO Total Wellbeing Diet had the edge in weight loss over a high-carbohydrate/low-fat diet…
Weight loss edge? What weight loss edge? According to the published research paper, weight loss was 7.3 kg plus or minus 0.3 kg on both the high red meat diet and the control diet, the high carb-low/fat diet. There was no weight loss edge after 12 weeks.
What about after 12 months?
Last year, about three years after the TWD was published, and about two years after it had sold enough copies to destroy a small forest, Clifton et al published the results of the 12 month followup study. Note, this study was complete before the first TWD book was published, but the results were only published last year … and without the usual press release and fan fare.
What happened to the women in his study during the 12 months after the initial 12 week study? On, average, they started putting the weight back on and by the end of 12 months, there was still no statistically significant weight loss edge between his diet and the control diet. Don’t take my word for it, check.
But reading the full paper shows you the problems are even worse. Remember that Houston quotes Clifton criticising population studies. But Clifton also praises controlled studies because they can handle confounders better. In his controlled study, women were allocated to the two diets randomly. This is precisely to control for unknown confounders. Quite right too. But at the end of 12 months, so few people had stuck to the TWD that Clifton couldn’t analyse the data in the obvious way. The only way Clifton could dredge out a positive spin was to ungroup his research subjects … in effect losing his confounder control. This is the mess behind the CSIRO’s boast that its diet is scientifically proven.
Neither Clifton, Olver, Truswell or Droulez mount a single telling argument against the attribution of causality by the WCRF to red meat. As I stated in my article last year, this attribution of causality together with data from the Victorian Cancer Council indicates that 6000 new cases of bowel cancer each year are due to eating more than one red meat serving per week.
Which brings us to Stewart Truswell. If you look at page 62 of the current National Health and Medical Research Council Dietary Guidelines for Australian Adults, you will find red meat exonerated from any role in colon cancer with reference to an “expert panel”. What NHMRC didn’t tell dietary guideline readers is that this panel was set up by MLA who paid the chairman … guess who? Stewart Truswell.
After the last WCRF report put a gentle finger on meat in 1997, MLA put together a damage control panel in 1998 and Truswell was the paid chair. So, in its 2003 Guidelines, NHMRC effectively ignored another five years worth of research and went with that panel’s opinion … did I mention Truswell was on the writing committee of the guidelines? You don’t get to be an elder statesman for nothing.
Now, if the WCRF report with its 150+ scientific authors is correct, then there are 6000 new cases of this cancer which are avoidable in Australia each year. I’ve had a good friend die of bowel cancer and its not a good way to die. So it’s hard to exaggerate my disgust that anyone, let alone Truswell, could call this a storm in a teacup. The source which exposes Truswell’s MLA payments is the British Medical Journal,
You won’t be able to check this however, unless you are a subscriber.