What follows is my entire correspondence with Robert Cohen. It’s copied here verbatim, except for footers, full headers, and repetitively quoted messages. Because of an unnoticed clock error on my computer, the dates for several of the first messages appear out of sequence. But the messages are displayed in the order in which they were sent and received. (NOTE: A glitch has caused the lines to become double-spaced.)

From: "Syd Baumel" <baumel@mts.net>

To: <notmilkman@notmilk.com>

Subject: About Hjartaker et al.

Date: Mon, 11 Mar 2002 21:47:58 -0600

Dear Robert,

I'm doing a report for aquarianonline.com on the controversy over milk.

It's a larger version of a print article I wrote for plant-based, an outreach publication 

within The Aquarian from the Winnipeg Vegetarian Association, of which I'm a member

(I'm a vegan). I'm also editor of The Aquarian.

I'm aware of the controversy over your critique of Lynette [sic: Anette] Hjartaker et

al.'s study suggesting milk protects against breast cancer. I've read

Stephen Walsh's open letter and corresponded with him and with Dr.

Hjartaker. I've also read the original paper. I'd like to understand why

you made the contested interpretations so I can report on this issue

fairly and accurately. May I ask you a few questions? I'll quote from

your critique and preface my queries with ****.

You wrote:

Here are my criticisms of the "study."

1) 317 of the 48,844 women in the study got breast cancer (six tenths of

one percent), but the study actually began with 57,664 women. Why were

the data from 8820 women eliminated? It turns out that 986 of those

women had cancer too (11%). What does that indicate regarding the entire

study?

****Do you have any reason to reject Hjartaker's explanation (and

Walsh's and my own impression as a health/medicine writer) that this was

a result of standard procedure in a prospective cohort study. Here's

what she wrote to me:

It is correct that we excluded 986 women with cancer prior to the study

onset. This is in line with the study design we used, that is, a cohort

study. An inclusion criteria for participating in this kind of study is

that the subject has to be free from the disease under investigation

when

included. That is, cohort studies always only include subjects that are

free from the disease under investigation. In our situation, free from

cancer.

Including only subjects free from the disease under investigation gives

cohort studies a stronger design than for instance case-control studies

(which include both diseased and healthy subjects).

One reason is that subjects who have got a disease often will report

differently on various exposure variables (such as dietary habits) than

subjects who have not got the disease. Diseased subjects are more likely

to

thoroughly examined their dietary habits etc.

You can read about this in e.g. Rothman and Greenlands's book 'Moderne

epidemiology'.

****If you still think there is something fishy about this, can you

answer your own question: "What does that indicate regarding the entire

study?"

3) TABLE 3 reveals the incidence rate ratios of breast cancer according

to milk consumption as a child and as an adult.

Based upon population statistics supplied by the authors, the

expectatiopn of breast cancers for low milk consuming females was 156

cases out of 311. The actual number of cases was only 42.

****Going by the case data you give, you must have meant the

age-adjusted model in Table 4. I can't tell from Hjartaker's paper which

"population statistics supplied by the authors" you used to arrive at

your expected case values, which seem radically inconsistent with the

milk consumption figures in Table 1. Could you provide these to me?

Here's what Hjartaker wrote when I asked her to comment on your figures:

It's right [i.e. Walsh's analysis], - Cohen must have made a gross and

arbitrary exaggeration of

the proportion of women in the low milk intake group.

The analyses we applied was Cox regression. In these analyses you don't

use

number of persons in the different categories, but rather person-years,

and calculate RATE RATIOS. In that case, one takes the follow-up time

into

account as well as the number of people included, and by that calculates

'number of person years'. Secondly, we adjust for several variables as

described in our analyses.

Taking both the follow-up time and the adjustments into account one will

not get exactly the same figures as Walsh has reached, but for this

purpose

the figures should be close enough.

____

Thanks for your consideration.

Syd Baumel

Editor, The Aquarian

www.aquarianonline.com

_______________________________________________________

Subject: About Hjartaker et al.

From: "i4crob@earthlink.net" <i4crob@earthlink.net>

Date: Thu, 14 Mar 2002 04:31:29 -0500

To: "baumel@mts.net" <baumel@mts.net>

Dear Syd,

Please obtain an ORIGINAL copy of the study,

and I will guide you through it.

You will note that the levels of milk consumption

change within the study (low, medium,

high doses change from 1-3 to 1-5, for example).

This is an extremely biased and poorly designed

study, and I will be happy to be your guide.

When you are ready to discuss same, call my office

at 201-871-5871.

Regards,

Robert

________________________________________________________

From: "Syd Baumel" <baumel@mts.net>

To: <i4crob@earthlink.net>

Subject: Re: About Hjartaker et al.

Date: Wed, 13 Mar 2002 13:54:00 -0600

Hi Robert,

I do have a copy of the original paper (as I wrote), i.e. one I

photocopied at the local medical library and have read from start to

finish (well, minus some of the really boring parts). Would it be asking

too much, as a favour to an overworked writer with a less than iron-clad

grasp of Cox proportional hazards regression analyses, if you could just

spell out in writing the details of your critique? I have years of

experience reviewing medical literature, so if a scientist can

understand your critique, so probably can I, with a little homework,

perhaps. If not, I'll run it by Walsh, Hjartaker, or some more neutral

third party. Indeed, even if I do follow it perfectly I would like to

run it by them for comment and return those comments to you for review

or rebuttal. I gather you relish the opportunity to debate milk-related

issues, and this would be a good way to accomplish that. Also, there

would be no danger of anyone being misquoted or taken out of context.

Even better (I'm on a roll here), how about if we cc the correspondence

to Walsh, Hjartaker, and anyone else you might like to include. I

suspect Walsh would welcome the opportunity to cc it to the IVU-SCI

newsgroup as well, and it could also be "simulcast" to your newsgroup,

if you like. Let's open this issue up to public discussion and review.

What do you think?

Thanks,

Syd

_______________________________________________________

Subject: About Hjartaker et al.

From: "i4crob@earthlink.net" <i4crob@earthlink.net>

Date: Thu, 14 Mar 2002 16:44:13 -0500

To: "baumel@mts.net" <baumel@mts.net>

CC: "dorietz@awod.com" <dorietz@awod.com>

baumel@mts.net wrote

{I do have a copy of the original paper (as I wrote), i.e. one I

photocopied at the local medical library and have read from start to

finish (well, minus some of the really boring parts). Would it be asking

too much, as a favour to an overworked writer with a less than iron-clad

grasp of Cox proportional hazards regression analyses, if you could just

spell out in writing the details of your critique?}

Huh?

I've already done that.

LOL

Yes, it would be a bit too much. I am on the way

to Toronto. Packing my van. Sorry you did not take

advantage of my phone call offer. Perhaps some other time.

{I have years of experience reviewing medical literature, so 

if a scientist can understand your critique, so probably can 

I, with a little homework, perhaps. If not, I'll run it by 

Walsh, Hjartaker, or some more neutral third party.}

Indeed, even if I do follow it perfectly I would like to

run it by them for comment and return those comments to 

you for review or rebuttal.} 

Perhaps you are not familiar with my experiences with

Walsh, but I will have nothing to do with him.

I am not interested in conversing with him. Nor do

I have any interest in anything he has to say. End

of subject.

{Also, there would be no danger of anyone being misquoted 

or taken out of context. Even better (I'm on a roll here), 

how about if we cc the correspondence to Walsh, Hjartaker, 

and anyone else you might like to include. I suspect Walsh 

would welcome the opportunity to cc it to the IVU-SCI

newsgroup as well, and it could also be "simulcast" to 

your newsgroup, if you like. Let's open this issue up to 

public discussion and review.

What do you think?} 

What do I think, Sid?

LOL

Judging by the past, I think that you may very well

cut and paste my comments together, just as others

have done during this recent controversy.

You are invited to post detailed comments to my guestbook.

So is Dr. Hjartaker. Mr. Walsh may take his forum anywhere 

else he wishes, but I will have nothing to do with him.

Robert Cohen

http://www.notmilk.com

_______________________________________________

From: "Syd Baumel" <baumel@mts.net>

To: <i4crob@earthlink.net>

Subject: Re: About Hjartaker et al.

Date: Sun, 24 Mar 2002 00:52:02 -0600

Robert,

May I call you then to discuss the Hjartaker paper? I'd like to record

the interview for accuracy (as I always do), if you have no objections.

Would an afternoon or evening call (CST) work for you?

Syd

_______________________________________________

Subject: About Hjartaker et al.

From: "i4crob@earthlink.net" <i4crob@earthlink.net>

Date: Mon, 25 Mar 2002 09:38:28 -0500

To: "baumel@mts.net" <baumel@mts.net>

Robert,

May I call you then to discuss the Hjartaker paper? I'd like to record

the interview for accuracy (as I always do), if you have no objections.

Would an afternoon or evening call (CST) work for you?

Dear Syd,

I look forward to it! Please call Lisa at

201-871-5871. She schedules all interviews 

for me. Save me a bit of time please, as I

place about four cubic feet of papers into

storage each week, and FAX the original

Hjartaker study to my office. 

FAX: 201-871-9304

Regards,

Robert Cohen

________________________________________________________

From: "Syd Baumel" <baumel@mts.net>

To: <i4crob@earthlink.net>

Subject: Re: About Hjartaker et al.

Date: Sun, 24 Mar 2002 12:26:11 -0600

Robert,

Thanks. Could I ask you to dig up that paper yourself? I'd love to

accomodate you, but my photocopy is so reduced I have to squint to read

it and a fax will be even less legible - plus I don't have a fax machine

anyway! (just my computer, which requires pre-scanning), so I'd have to

"outsource" to fax it to you.

If your copy has really gone hiding, let me know and I'll see what I can

do. Please let me know either way so I can schedule that call.

Regards,

Syd

___________________________________________________

Subject: About Hjartaker et al.

From: "i4crob@earthlink.net" <i4crob@earthlink.net>

Date: Mon, 25 Mar 2002 14:43:09 -0500

To: "baumel@mts.net" <baumel@mts.net>

Robert,

Thanks. Could I ask you to dig up that paper yourself? 

You could, but it will take a few weeks.

I am on tour, and last week I transported

about 150 boxes worth of "stuff" into a

monthly storage shed. I should have a better

storage system, but I do not. Why not find a good

copy machine?

My snail mail:

Robert Cohen

560 Oradell Avenue

Oradell, NJ 07649

_____________________________________________________

From: "Syd Baumel" <baumel@mts.net>

To: <i4crob@earthlink.net>

Subject: Re: About Hjartaker et al.

Date: Sun, 24 Mar 2002 16:31:53 -0600

Robert,

I'm trying to meet a commitment to complete and publish the report

online by the end of the month. So it would be highly desirable if I

could talk to you this week. How about if I scan and email you the study

as printable (or viewable) gifs, jpegs, or tifs (what's your preference?

gifs seem the easiest for me)? The file size (complete) could be as low

as 1 Mb.

Syd

________________________________________

Subject: About Hjartaker et al.

From: "i4crob@earthlink.net" <i4crob@earthlink.net>

Date: Mon, 25 Mar 2002 18:02:09 -0500

To: "baumel@mts.net" <baumel@mts.net>

I'm trying to meet a commitment to complete and publish the report

online by the end of the month. So it would be highly desirable if I

could talk to you this week. 

That is not possible.

I'm trying to meet a commitment to complete 

and publish the report online by the end of 

the month. 

I have three more hours of radio interviews today.

I have a few dozen interviews during the next week,

three speaking engagements, a number of meetings

in New York city. I will be in Los Angeles from

Wednesday until Friday. I have a few TV shows to do,

and am doing an informercial. I also have an intense

writing schedule. I receive over 2000 email letters

per day, with dozens of daily requests for interviews.

I have advised that you speak with Lisa so that

we can work out a time. to schedule your interview.

I do NOT open up files on the Internet. I make no

exceptions. My FAX number is 201-871-9304.

Regards,

Robert Cohen

_______________________________________________

From: "Syd Baumel" <baumel@mts.net>

To: <i4crob@earthlink.net>

Subject: Hjartaker et al.

Date: Tue, 16 Apr 2002 22:36:58 -0500

Hi Robert,

I thought I'd run this by you in advance of our scheduled conversation

next Tuesday. It's from a draft of my article-in-progress on the

conflict over the Hjartaker study (which I faxed to your office shortly

after your last email; your secretary confirmed receipt when we

scheduled the interview). This part focuses on the two major areas of

disagreement between you and Walsh and Hjartaker. I welcome your

comments.

Regards,

Syd Baumel

www.aquarianonline.com

____

Based on what he claimed to be a truly critical analysis of the original

paper, Cohen charged the authors with two particularly grievous crimes

of deceit.

Insinuating shady data manipulation, Cohen wrote:

"317 of the 48,844 women in the study got breast cancer (six tenths of

one percent), but the study actually began with 57,664 women. Why were

the data from 8820 women eliminated? It turns out that 986 of those

women had cancer too (11%). What does that indicate regarding the entire

study?"

In answer to Cohen's finger-pointing, Walsh explained that the 986

women, all of whom were found to have a prior diagnosis of cancer at the

time they enrolled in the study, were automatically excluded by virtue

of standard, sound study design. The paper, which I have also read,

detailed the equally innocuous reasons for excluding the remainder of

the 8820 women. Most, for example, had simply not been asked to fill out

a dietary questionnaire when they were entered into the study.

As Hjartåker (whom I asked for comment) explains: "Cohort studies

[studies that follow a clearly defined group of people over time] always

only include subjects that are free from the disease under

investigation. In our situation, free from cancer."

Indeed, the study was essentially a fishing expedition to take the

measure (dietary and otherwise) of a very large, randomly selected

"cohort" of cancer-free premenopausal women at its inception in 1991 and

1992, and, as the years go by, to see how those measurements relate to

who gets cancer and who doesn't.

. . . . .

When I emailed Hjartåker for her two cents worth, she agreed that "Cohen

must have made a gross and arbitrary exaggeration of the proportion of

women in the low milk intake group." Her group had used "Cox

regression," she explained. "In these analyses you don't use number of

persons in the different categories, but rather person-years, and

calculate RATE RATIOS. . . .Taking both the follow-up time and the

adjustments into account one will not get exactly the same figures as

Walsh has reached," Hjartåker wrote, "but for this purpose the figures

should be close enough."

I wanted to do better than "close enough," so I asked Hjartåker if she

could provide the details of how she and her associates had converted

the published figures of four discrete categories of milk intake into

the three categories used in the combined analysis.

"I hope this doesn't seem like an 'interrogation,'" I wrote. "My purpose

is to present the truth as transparently and unambiguously as possible."

That was March 17, and while I was waiting for Dr. Hjartåker's reply,

Walsh steered me to a few lines in the "methods" section I had

bumblingly overlooked wherein Hjartåker and company do explain their

formula. It is a bit of a braintwister, which might explain Cohen's

incomprehensible interpretation. To quote from the paper:

"The combined effect of childhood and adult milk consumption was

examined by constructing a 3-level variable: 'low consumption,' defined

as no milk consumption on at least one of the occasions and not more

than next-lowest consumption on the other occasion; 'high consumption,'

defined as the highest milk consumption on at least 1 of the occasions

and not less than the next-highest consumption on the other occasions

and 'moderate consumption,' defined as all the other combinations."

Judging by the raw data in the paper alone, this little algorithm meant

that the low group would have contained somewhere between 6.6% and 16.1%

of the women, the high group between 1.9% and 12.3%, and the moderate

group everyone else. Taking those ranges and comparing them to the

relative risks that the researchers published, but not knowing about the

inclusion of person-years in the calculations, Walsh deduced that low,

moderate, and high had been 11%, 82%, and 7%, respectively. When

Hjartaker later responded to my query with the raw "person years" in

each group (low: 32 280; Moderate: 242 209; High: 22 271) in the

age-adjusted analysis, they almost perfectly matched Walsh's estimate:

10.9%, 81.6%, and 7.5%. Most importantly, they weren't even remotely

similar to the figures Cohen had invoked to arrive at his damning

conclusions of fraud and incompetence. When I used these figures

together with the published numbers of breast cancer cases (adjusted

for age) in the three different groups to calculate the RRs (relative

risks), I obtained the following results:

Low milk consumption: 42 cases (42/32,280 person years = 0.130%)

Moderate milk consumption: 254 cases (254/242,209 = 0.1049%)

High milk consumption: 15 cases (15/22,271 = 0.0674%)

Setting "Low" (0.130%) as RR = 1, as Hjartaker et al. did, yielded the

following RRs for the other two groups:

Moderate: 0.1049/0.1300 = 0.807 (Hjartaker et al.'s published figure:

0.81)

High: 0.0674/0.1300 = 0.518 (Hjartaker et al.: 0.54)

That's close enough for me, and if Robert Cohen or anyone else can still

find fault with Hjartaker et al.'s data, I'm all ears.

__________________________________________________

Subject: Hjartaker et al.

From: "i4crob@earthlink.net" <i4crob@earthlink.net>

Date: Wed, 17 Apr 2002 10:46:08 -0400

To: "baumel@mts.net" <baumel@mts.net>

CC: "notmilk@earthlink.net" <notmilk@earthlink.net>

In any study, criteria must remain standardized.

Let me ask you this.

In th AH study, please define:

CHILDREN

Low milk consumption - How much?

Medium consumption - How much?

High consumption - How much?

ADULT

Low milk consumption - How much?

Medium consumption - How much?

High consumption - How much?

Thanks,

Robert

________________________________________________

Subject: Hjartaker et al.

From: "i4crob@earthlink.net" <i4crob@earthlink.net>

Date: Wed, 17 Apr 2002 11:09:28 -0400

To: "baumel@mts.net" <baumel@mts.net>

CC: "notmilk@earthlink.net" <notmilk@earthlink.net>

I thought I'd run this by you in advance of our scheduled conversation

next Tuesday. I welcome your comments.

Dear Syd,

Is the purpose of scientific pursuit the discovery of TRUTH,

or the manipulation of data. One can make an argument that

an individual drowned in a stream with the average depth pf

four inches. The stream could have an average depth of one inch 

for 100 square miles, but that lethal 10 squar foot section

could be the key event.

That which you have written contains

biased language. You wrote:

<<Insinuating shady data manipulation, Cohen wrote>>

Such language not only biases your potential readers, but

shows me your intent.

You continue:

<<In answer to Cohen's finger-pointing>>

So, up to this point you have me acting shady, insinuating,

and finger pointing. You then have my critic say:

<<986 women, all of whom were found to have a prior diagnosis of cancer at the time they enrolled in the study, were auto=

matically excluded by virtue

of standard, sound study design.>>

This is NOT standard study design. Again, your bias.

If the purpose of the study is to determine how

many people develop cancer as a result of childhood

habits, one must include the entire population from

childhood to adulthood. By eliminating those with cancer,

one biases the study.

You write:

The paper, which I have also read,

detailed the equally innocuous reasons for 

excluding the remainder of the 8820 women. 

Most, for example, had simply not been asked to fill out

a dietary questionnaire when they were entered into the study.>>

"Simply not been asked to fill out a dietary 

questionairre??????"

Why not? More experimental bias, that's why.

<<As Hjart=E5ker (whom I asked for comment) explains: "Cohort studies

[studies that follow a clearly defined group of people over time] always

only include subjects that are free from the disease under

investigation. In our situation, free from cancer.">>

That reveals the flaw. Hjartaker clearly set up the parameter

of the study. She defined the terms. Does milk consumption 

as a child cause cancer as an adult? This question cannot be answered

honestly by eliminating those who got cancer.

cancer?

<<Indeed, the study was essentially a fishing expedition to take the

measure (dietary and otherwise) of a very large, randomly selected

"cohort" of cancer-free premenopausal women at its inception in 1991 and

1992, and, as the years go by, to see how those measurements relate to

who gets cancer and who doesn't.>>

Indeed?????

Essentially???

They were not randomly selected if a large part of the population

was excused for having cancer which they may have gotten as a result

of childhood milk consumption.

<<When I emailed Hjart=E5ker for her two cents worth, she agreed that "Cohen

must have made a gross and arbitrary exaggeration of the proportion of

women in the low milk intake group.">>

ASK HER HOW SHE DETERMINED LOW, MEDIUM, AND HIGH CONSUMPTION.

The numbers are different for kids and adults. How did

she draw her arbitrary lines. Is low consumption

1-2 glasses or less than one, etc.

<<Walsh steered me to a few lines in the "methods" section I had

bumblingly overlooked wherein Hjart=E5ker and company do explain their

formula. It is a bit of a braintwister, which might explain Cohen's

incomprehensible interpretation. To quote from the paper:>>

Incomprehensible intermpretation?

Perhaps we should not do the interview, and have you just write whatver you will.

____________________________________________________

From: "Syd Baumel" <baumel@mts.net>

To: <i4crob@earthlink.net>

Subject: Re: Hjartaker et al.

Date: Wed, 17 Apr 2002 14:48:08 -0500

----- Original Message -----

From: <i4crob@earthlink.net>

To: <baumel@mts.net>

Cc: <notmilk@earthlink.net>

Sent: Wednesday, April 17, 2002 10:09 AM

Subject: Hjartaker et al.
 
 

I thought I'd run this by you in advance of our scheduled conversation

next Tuesday. I welcome your comments.

Dear Syd,

Is the purpose of scientific pursuit the discovery of TRUTH,

or the manipulation of data. One can make an argument that

an individual drowned in a stream with the average depth pf

four inches. The stream could have an average depth of one inch

for 100 square miles, but that lethal 10 squar foot section

could be the key event.

That which you have written contains

biased language. You wrote:

<<Insinuating shady data manipulation, Cohen wrote>>

Such language not only biases your potential readers, but

shows me your intent.
 
 

***Robert, those judgemental words (far, far less judgemental than your

published words with regard to Hjartaker et al.) represent how I regard

your position based on my current knowledge. I'd be more than happy to

sing your praises instead if our further communications give me sound

reasons to. Please remember, this is just a draft of my article. I could

have - and had wanted to, because of scheduling commitments - published

it weeks ago after you refused to discuss this subject by email. But I

decided I had a responsibility not to go to print until I'd given you

every reasonable opportunity to explain your views to me in the manner

of your choosing. I even went to considerable trouble to scan, format,

and fax the original paper to you to spare you the trouble and delay of

having to retrieve your copy, as you wrote to me, from boxes upon boxes

of stored and uncatalogued material.

***Also, let me ask you: were you NOT, then, insinuating the researchers

had manipulated data? You'd already described this as "the fraudulent

study of the century" and primed your readers expectations by writing:

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

deceit. This lie made every major American newspaper and television news

program." How else did you want readers to interpret your reference to

all those excluded cancer cases? I would really like to know if you had

something else in mind, because I'll be happy to quote you on that.

You continue:

<<In answer to Cohen's finger-pointing>>

So, up to this point you have me acting shady, insinuating,

and finger pointing. You then have my critic say:

***Again, if you deserve credit for exposing scientific fraud rather

than criticism for an unfounded attack on people conducting proper

scientific research, I will very happily report that. Just convince me.

I take my responsibilities as a journalist very seriously.

<<986 women, all of whom were found to have a prior diagnosis of cancer

at the time they enrolled in the study, were automatically excluded by

virtue

of standard, sound study design.>>

This is NOT standard study design. Again, your bias.

If the purpose of the study is to determine how

many people develop cancer as a result of childhood

habits, one must include the entire population from

childhood to adulthood. By eliminating those with cancer,

one biases the study.

****If the reseachers had set out to answer that question via

case-control methodology, you'd have a point. But they chose to try and

answer it via what is generally, if not universally, accepted by

scientists as a less error-prone design: a prospective cohort study. For

that reason, those women were not even enrolled in the study (i.e. they

failed to meet the a priori inclusion criteria), let alone given dietary

questionnaires and then - as I believe most readers who trust your

critique would conclude - swept under the rug to bury some anti-milk

effect. Again, if your rhetorical intent was otherwise, please say

exactly what you want(ed) your readers to think.

You write:

The paper, which I have also read,

detailed the equally innocuous reasons for

excluding the remainder of the 8820 women.

Most, for example, had simply not been asked to fill out

a dietary questionnaire when they were entered into the study.>>

"Simply not been asked to fill out a dietary

questionairre??????"

Why not? More experimental bias, that's why.
 
 

****Bias in which direction? Anti-milk or pro-milk? According to the

paper (p. 888), "In 1991-92, a random, nationwide sample of 100,000

Norwegian women born 1943-57 were drawn from the National Register and

invited to participate in the Norwegian Women and Cancer Study (NOWAC).

A total of 61,000 women were randomly sampled in 1991, and an additional

39,000 (Norwegian citizens only) in 1992. The women received a mailed

letter of invitation requesting informed consent and a self-instructive

questionnaire." Later (p. 889): "Of the 100,000 women initially invited

to participate in NOWAC, 6000 were given a questionnaire without dietary

questions. The responders of this questionnaire (n = 3,694) were not

included in the present analyses." I'm at a loss to understand how a

potentially biased group of researchers could have used this to their

advantage, e.g. how they could have known in advance that the dietary

habits of these virtually anonymous 6000 women would later correlate

with first cases of breast cancer in a way that would incriminate milk.

If I'm missing something here, please spell it out.

<<As Hjartåker (whom I asked for comment) explains: "Cohort studies

[studies that follow a clearly defined group of people over time] always

only include subjects that are free from the disease under

investigation. In our situation, free from cancer.">>

That reveals the flaw. Hjartaker clearly set up the parameter

of the study. She defined the terms. Does milk consumption

as a child cause cancer as an adult? This question cannot be answered

honestly by eliminating those who got cancer.

cancer?

***Again, a prospective cohort study, which is one of a handful of

standard research designs for studying people and disease, BY DEFINITION

must not include subjects who already have (or may recently have had)

the disease being investigated (or who have ANY serious disease in

hypercautious cohort studies). This is standard textbook stuff - not

some arbitrary or biased decision by Hjartaker who would have shown

professional incompetence if she had designed the study otherwise, as

you suggest she should have. The idea of a prospective design is to

start with a blank slate, "no baggage." Also, please note that

Hjartaker's study was a "study within a study" (like so many other such

studies based on The Nurses' Health Study, the Honolulu Heart Program,

the Framingham Study, etc.), in this case the NOWAC study which was

formulated to look for causes of ALL types of cancer in women by

comparing baseline dietary and other characteristics in women who went

on to get cancer and those who didn't. This is why women with any type

of cancer who responded to the invitation to join the study in 1991-92

were told, "sorry, no thanks." Rightly or wrongly, scientists believe

that if you already have a disease like breast cancer, simply knowing

that could bias your recollection or perception of the very dietary and

other lifestyle factors under study. For example, you might well think

"I got this because I drank too much milk"; and that suspicion could

lead you to overestimate how much milk you actually drank. Or, knowing

that you have breast cancer, you might have changed your milk-drinking

habits ("from now on, no cow hormones for me, thank you very much"). If

you were included in a prospective study of this kind which asks (to

quote from the paper) about your "adult milk consumption at baseline"

(i.e. when you received the questionnaire), you would report a low

consumption, but it would have been as a RESULT of your diagnosis of

breast cancer not as a cause.

<<Indeed, the study was essentially a fishing expedition to take the

measure (dietary and otherwise) of a very large, randomly selected

"cohort" of cancer-free premenopausal women at its inception in 1991 and

1992, and, as the years go by, to see how those measurements relate to

who gets cancer and who doesn't.>>

Indeed?????

Essentially???

They were not randomly selected if a large part of the population

was excused for having cancer which they may have gotten as a result

of childhood milk consumption.

****Or, if one approaches this question without bias, who may have

gotten cancer as a result of having drunk relatively LESS milk in

childhood. There was no way for the researchers to know the answer to

that question in advance. Instead of trying to answer it by means of a

comparatively bias- and error-prone case-cohort design - which would

have allowed the inclusion of the women with breast cancer at the

outset - the researchers opted to investigate the possible links between

childhood and adulthood diet (milk was just one of many items on the

dietary questionnaire) and breast cancer through a LESS biased study

design: a prospective cohort study. As you know, researchers have

limited budgets. It would have been interesting to have seen what a

case-cohort study would have shown; but one can hardly fault the

researchers for piling their pennies into a more objective prospective

study instead.

<<When I emailed Hjartåker for her two cents worth, she agreed that

"Cohen

must have made a gross and arbitrary exaggeration of the proportion of

women in the low milk intake group.">>

ASK HER HOW SHE DETERMINED LOW, MEDIUM, AND HIGH CONSUMPTION.

The numbers are different for kids and adults. How did

she draw her arbitrary lines. Is low consumption

1-2 glasses or less than one, etc.

<<Walsh steered me to a few lines in the "methods" section I had

bumblingly overlooked wherein Hjartåker and company do explain their

formula. It is a bit of a braintwister, which might explain Cohen's

incomprehensible interpretation. To quote from the paper:>>

****The explanation, as I reprinted from Hjartaker's paper itself and

explained in the draft I sent to you, makes it very clear how they

created these groups. If there's something about their method that you

think could have biased their results, please explain.

Incomprehensible intermpretation?

Perhaps we should not do the interview, and have you just write whatver

you will.

****That's entirely up to you if you feel you have nothing more to add

to explain how you arrived at your interpretations, calculations, and

conclusions, particularly this one:

. . . the expectatiopn [sic] of breast cancers for low milk consuming

females was 156 cases out of 311. The actual number of cases was only

42.

The expected number of cases of breast cancer for the moderate and high

milk consumption group was 155 cases. The actual number of cases of

breast cancer for the milk drinkers was 269.

In other words, the authors mis-read [sic] their own data.

Women who drank a lot of milk as children developed more cases of breast

cancer than notmilk users. How much more? A factor of 640%!

****Please let me know if you'd still like to speak to me on April 23,

as scheduled. Otherwise, I'd be happy to reprint this exchange as an

appendix to my article and let readers decide for themselves where the

bias lies.

*****In the interest of being thorough, let me also reply to your other

message here. You wrote:

In any study, criteria must remain standardized.

Let me ask you this.

In th AH study, please define:

CHILDREN

Low milk consumption - How much?

Medium consumption - How much?

High consumption - How much?

ADULT

Low milk consumption - How much?

Medium consumption - How much?

High consumption - How much?

Thanks,

Robert

****As you know, the questionnaire gave the women a four part multiple

choice answer for the two separate questions about their childhood and

adulthood levels of milk consumption, respectively. As you should also

know, having read this paper so critically, the researchers write that

"Based on wholesale statistics, we know that the per capita consumption

of milk in Norway in the 1950s, when the women were growing up, was

high." Acc0rdingly, the multiple choice for childhood milk consumption

was set at higher levels than for adult milk consumption, but both

included "none" as the bottom level, and 7 or more glasses a day as the

highest level for childhood and more than 3 glasses a day as the highest

level for adulthood. I don't see any room for an anti-milk bias in this

decision, which also seems like a reasonable accomodation to the fact

that children tend to drink more milk so they can "grow up and be

strong" than adults.

****If you can provide a credible argument as to how this lack of

standardization between the multiple choices for childhood and adulthood

milk consumption might have introduced significant bias FOR MILK into

the study, I'm all ears.

Regards,

Syd Baumel

______________________________________________________

Subject: LET'S TRY THIS AGAIN, SYD

From: "i4crob@earthlink.net" <i4crob@earthlink.net>

Date: Wed, 17 Apr 2002 21:14:52 -0400

To: "baumel@mts.net" <baumel@mts.net>

Dear Syd,

Perhaps you do better with "fill in the blanks"

CHILDREN 

Low milk consumption - How much? ______

Medium consumption - How much? ______

High consumption - How much? ______ 

ADULT 

Low milk consumption - How much? ______

Medium consumption - How much? ______

High consumption - How much? ______

Are you capable of keeping this simple?

;>)

Thanks, 

Robert 

______________________________________________

From: "Syd Baumel" <baumel@mts.net>

To: <i4crob@earthlink.net>

Subject: Re: LET'S TRY THIS AGAIN, SYD

Date: Thu, 18 Apr 2002 01:11:07 -0500

----- Original Message -----

From: <i4crob@earthlink.net>

To: <baumel@mts.net>

Sent: Wednesday, April 17, 2002 8:14 PM

Subject: LET'S TRY THIS AGAIN, SYD
 
 

Dear Syd,

Perhaps you do better with "fill in the blanks"

CHILDREN

Low milk consumption - How much? ______

Medium consumption - How much? ______

High consumption - How much? ______

ADULT

Low milk consumption - How much? ______

Medium consumption - How much? ______

High consumption - How much? ______

Are you capable of keeping this simple?

;>)

****Robert, I'm keeping it as simple as I can without being inaccurately

simplistic. I've gone to a lot of trouble to explain myself as clearly

and thoroughly as I can. If there's anything I've written that you don't

understand, please say so, and I'll try again.

***To answer your question: As someone who has read the original paper

and (I hope!) my previous emails, you should know that the questionnaire

didn't ask about "low," "moderate," or "high" consumption, but about

four quantitatively defined levels of consumption (childhood [glasses

per day]: none, 1-3, 4-6, 7 or more; adulthood: none, 0.1-1.0, 1.1-3.0,

3.1 or more). So the blanks in your question can't be filled - unless

you would like to fold two categories into one and reclassify the

remaining three as low, moderate, and high. Is there any way you believe

that would make this a better study?

***Perhaps you're hinting at the question of how to reclassify in an

unbiased way the women into three "low," "moderate," and "high" groups

that meaningfully combine their childhood AND adulthood (at entry into

the study) milk consumption. Hjartaker et al. chose to do it in a way

that strikes me as an elegant solution to the problem. Can you think of

any other way(s) that might have yielded a significantly different

result, in particular a result that would have been statistically

insignificant or significantly AGAINST milk?

***I hope I've answered your question. I'd still like to know how you

arrived at the risk ratio figures you published in your critique,

because this, of course, is key to understanding your contention that a

serious fraud has been committed by the researchers. Failing that

explanation, I have to go with what I've got, based on my own careful

reading of the paper and consultation with Stephen Walsh and Anette

Hjartaker. So again, please explain yourself and show me the error of my

ways or admit that you were wrong.

***Regards,

***Syd

Thanks,

Robert

_______________________________________________

Subject: FINAL ATTEMPT

From: "i4crob@earthlink.net" <i4crob@earthlink.net>

Date: Thu, 18 Apr 2002 05:20:09 -0400

To: "baumel@mts.net" <baumel@mts.net>

CC: "notmilk@earthlink.net" <notmilk@earthlink.net>

FINAL ATTEMPT

Dear Syd,

You are trying to play a ball game

with me (in the big leagues) and 

cannot even get to first base

(YOU NEVER PLAYED LITTLE LEAGUE BALL, 

DID YOU?)

You do not seem capable of answering 

even the simplest question without 

providing a long rambling statement.

This is my final request.

Failure to respond to this 

most basic question will

terminate our EMAIL 

communication, and end our interview.

PLEASE FILL IN THE BLANKS:

CHILDREN

Low milk consumption - How much? ______

Medium consumption - How much? ______

High consumption - How much? ______

ADULT

Low milk consumption - How much? ______

Medium consumption - How much? ______

High consumption - How much? ______

Robert Cohen

(Lisa--please file this. It's clear

that I am being set up, and this 

so-called reporter and his journal

may very well be hit with a libel 

and slander suit similar to the 

brief we are preparing for the

JN/ES thing. As per the other

EMAILS, this needs to be well

documented--Robert)

suit

________________________________________________

From: "Syd Baumel" <baumel@mts.net>

To: <i4crob@earthlink.net>

Subject: Re: FINAL ATTEMPT

Date: Thu, 18 Apr 2002 04:41:40 -0500

Robert,

You're right about one thing: I'm not in the same "big league" as you.

Thank God.

I have no idea how to play whatever game it is you think you're playing

or how to answer your riddle.

I'm happy to join you in terminating this fruitless correspondence.

Happy litigating. If you want to sue, sue me personally.

Syd Baumel

__________________________________

Subject: FINAL ATTEMPT

From: "i4crob@earthlink.net" <i4crob@earthlink.net>

Date: Thu, 18 Apr 2002 05:50:15 -0400

To: "baumel@mts.net" <baumel@mts.net>

CC: "notmilk@earthlink.net" <notmilk@earthlink.net>

You're right about one thing: I'm not in the same "big league" as you.

Thank God. 

Your REFUSAL to answer my repeated

request reveals your intentions, Syd.

I suspect that you tried, and in doing so

rervealed the basic flaw of this study.

Now that you are armed with truth, I 

expect you to write it.

Failure to do so will respond in legal action

against you and your journal.

As you know, I have taken a lot of crap, and

will accept no more.

Should you wish to continue this discussion,

you must first respond to my SIMPLE

question.

I suspect that you will not.

If that is the case, you are a pathetic

excuse for a journalist.

Robert Cohen

______________________________

From: "Syd Baumel" <baumel@mts.net>

To: <i4crob@earthlink.net>

Subject: Re: FINAL ATTEMPT

Date: Thu, 18 Apr 2002 12:30:14 -0500

Robert,

I can't reveal something I don't know and that you refuse to tell me.

For the last time, explain yourself (as I have), or forever hold your

peace (such as it is).

Syd

(Anna: Please archive these messages, because, as you know, I know how

to write emails, but not how to store them and I have a nervous tic that

causes me to type whatever I'm saying to you into my emails to other

people. Also please see about that adjustment to my meds.)

_________________________________________

Subject: FINAL ATTEMPT

From: "i4crob@earthlink.net" <i4crob@earthlink.net>

Date: Thu, 18 Apr 2002 15:59:11 -0400

To: "baumel@mts.net" <baumel@mts.net>

I can't reveal something I don't know and that you refuse to tell me. 

Brush on, brush off, brush on, brush off, brush

on, brush off.

The information you seek is before you.

Unless you discover the secret, you will

not experience truth.

Read the AH study and fill in the blanks.

Your refusal to do so is perplexing.