By Syd Baumel

Over and over again in "Fooling Most of the People Most of the Time,"  Robert Cohen attempts to smear Anette Hjartåker and her associates – co-authors of the NOWAC breast cancer study – with a word he uses so much you wonder if he’s trying to tell us something about himself with it: bias.
  IS THAT REAL SCIENCE OR REAL BIAS?
 

The authors bring their biases to the discussion by writing . . .

  In his emails to me about the study, the "b" word also keeps coming up:
  This is an extremely biased and poorly designed study . . .

By eliminating those with cancer, one biases the study.

"Simply not been asked to fill out a dietary
questionairre??????"
Why not? More experimental bias, that's why.
 

Even stronger words – fraud and deceit – pepper Cohen’s "Fooling Most of the People" attack on the study:
 

Today I present you with an amazing story of scientific fraud and deceit.

 

THE FRAUDULENT STUDY OF THE CENTURY

Often times, the conclusions from published studies contradict their own data.

The "MILK PREVENTS BREAST CANCER" conclusion is one such example.

Shame on scientists for their deceit.

 

The intensity of Cohen’s mudslinging is exceeded only by the lack of evidence he marshalls to support it. It would, in fact, be remarkable if the kind of scientific incompetence and deceit Cohen attributes to the study’s authors would have passed peer review. Prior to its publication in the International Journal of Cancer, other scientists who specialize in nutritional epidemiology as well as the journal’s editors would have had ample opportunity to catch and correct (or reject) the kind of biased study design and misrepresented data that Cohen (uniquely) sees.

 

Unless . . . they too were in on the fraud.

 

For cynical critics of the medical-industrial complex, this is not an implausible scenario. 

Fortunately, there’s an easy way to check: Medline, the searchable index of virtually all published biomedical research.
 

I therefore searched Medline for any papers in the International Journal of Cancer involving dairy (keywords: "Int J Cancer"[Jour] dairy). I got 12 hits, in addition to the NOWAC study itself. All 12 had abstracts briefly summarizing their contents. One didn’t mention dairy, only a positive (adverse) association between saturated fat and bladder cancer. Quite likely the paper itself would show that fatty dairy foods contributed to that effect. Another abstract simply referred to a companion paper in the same issue on the relationship between dairy (and other animal products) and colon cancer. That companion paper, along with eight of the other ten studies, reported a positive (adverse) association between dairy consumption and various types of cancer in humans. Only one International Journal of Cancer study other than the Hjartåker paper had anything good to say about dairy: people with gliomas (a type of brain tumour) were less likely to have dairy and other allergies.

 

If the International Journal of Cancer has "a pro-dairy agenda," it’s doing a miserable job of it.

But what about Hjartåker herself and the two scientists who co-wrote the paper with her?
Scientists publish on an honour system that leaves many opportunities to suppress, spin, or otherwise massage their findings. They can even fabricate data; and some clearly do, as we know from those who occasionally get caught.
 

Again, a simple way to check if a biomedical scientist may be biased is to review their research record on Medline. If they consistently publish pro-dairy studies, for example, they may very well be biased. If, on the other hand, their research displays no consistent pattern for or against dairy, this suggests they're just doing their job.

I used Medline to check the record of all three authors of the NOWAC breast cancer study. The results were revealing; but not in a way that would give comfort to Robert Cohen.
 

The NOWAC study’s lead author, Anette Hjartåker, only had six studies on Medline – all of them published since the early 90s. Judging by the titles and abstracts, only one of these (excluding the NOWAC study) involved dairy. A 1997 case-control study, it found that people with thyroid cancer were more likely than healthy controls to report eating lots of butter and cheese.

So much for a Hjartåker pro-dairy agenda.

Also collaborating on the thyroid study was one of Hjartåker’s co-authors on the NOWAC study, E. Lund.

Could Lund be biased – an "infiltrator," perhaps, to use Cohen's lingo?

Searching Medline for E. Lund and "dairy" or E. Lund and "milk" yielded only the two studies with Hjartåker.

No agenda for Lund either.

The third author of the NOWAC study, P. Laake, had a bigger dairy record. He or she has published four other studies definitely or probably involving dairy. The abstract of the "probable" study reports no dietary connections of any kind to malignant melanoma. The other three studies are 2 to 1 "against" dairy:
 

• One suggests skim milk is protective against lung cancer, compared to whole milk.
• Another suggests skim milk, but not whole milk promotes prostate cancer.
• The third reports an increased risk for acute leukemia among dairy farmers.

Thus, P. Laake has co-authored two studies suggesting milk or dairy is good for you and two suggesting it’s bad. No agenda there either.

Hjartåker, Lund, Laake, and the International Journal of Cancer appear to be doing what scientists and medical journals are supposed to do: following the data wherever it leads and reporting it honestly so that the scientific community and the rest of us can separate fact from fiction about what’s good for us and what’s bad.

They could teach the Notmilkman a thing or two.