Lyon Diet Heart Study questions

[Paul Chefu]

The "Mediterranean" diet gets a lot of good notice for being
healthy overall, and especially good for the prevention of
heart disease. The Lyon Diet Heart Study is usually referenced
as the definitive finding on this type of diet.

Does anyone know if other studies have replicated the Lyon
findings? The reason I ask is that I went looking on the AHA
web site, and found a review of the study that raised a
significant questions about its validity.

http://www.americanheart.org/presenter.jhtml?identifier=4655

The passages that gave me pause were in the conclusions,:

************************************************
What were the problems with the study?

These results are quite impressive. However, limitations in
study methods raise questions about the true impact of this
diet on risk of recurrent heart disease and related measures.
Specifically, the baseline diet was only assessed in the
experimental group at the beginning of the study. The control
group's diet was only assessed at the conclusion. This was
done to avoid influencing the dietary behavior of these
subjects. Thus, it's not clear whether there were any dietary
changes made by the control group.

Dietary data at the final visit are reported for only 83 out
of 303 subjects (30 percent ) in the control group and 144 out
of 302 (less than 50 percent) in the experimental group. The
diet of the other subjects who completed the study is unknown.
This raises questions about the role of diet in explaining the
results reported for recurrent coronary events.
************************************************
They seem to be saying that the problems in both the design of
the study and the data gathering render its conclusions
suspect. If that is the case, what other well-designed studies
have been done to overcome those difficulties and make the
diet as universally recommendable as it is today?

Paul

[Jay Tanzma]

Paul Chefurka wrote:
>
> The "Mediterranean" diet gets a lot of good notice for being
> healthy overall, and especially good for the prevention of
> heart disease. The Lyon Diet Heart Study is usually
> referenced as the definitive finding on this type of diet.

It is not usually cited as definitive.

> Does anyone know if other studies have replicated the Lyon
> findings?

To the best of my knowledge, no other studies of the effect of
a Mediterranean in the secondary prevention of heart attacks
have been reported.

> The reason I ask is that I went looking on the AHA web site,
> and found a review of the study that raised a significant
> questions about its validity.
>
> http://www.americanheart.org/presenter.jhtml?identifier=4655
>
> The passages that gave me pause were in the conclusions,:
>
> ************************************************
> What were the problems with the study?
>
> These results are quite impressive.

The point estimate of the relative risk was quite impressive
(something like .3), but the study was stopped too soon, and
consequently, the confidence interval around the point
estimate was wide, with an upper limit > .8, IIRC. This does
not rule out the possibility that the intervention was only
moderately effective (or even not effective at all).

> However, limitations in study methods raise questions about
> the true impact of this diet on risk of recurrent heart
> disease and related measures. Specifically, the baseline
> diet was only assessed in the experimental group at the
> beginning of the study. The control group's diet was only
> assessed at the conclusion. This was done to avoid
> influencing the dietary behavior of these subjects. Thus,
> it's not clear whether there were any dietary changes made
> by the control group.

This is related to two of my criticisms of the study (which
are themselves related). One, the control and intervention
groups should have received equally strong deitary
intervention, and, two, the control group (IMO) should have
been put on a Step 2 diet. The rational for the first
criticism is that study subjects are known to respond
favorably from counselling or attention from investigators per
se. This "Hawthorne effect" should be controlled. The
rationale for the second criticism is twofold: everybody in
the control group should be on the same diet, not there own
self-selected diets, and two, that diet should be a
therapeutic one. The question being investigated should have
been, "Is the Mediterranean diet better than the currently
recommended therapeutic diet, not the habitual diet.

> Dietary data at the final visit are reported for only 83
> out of 303 subjects (30 percent ) in the control group and
> 144 out of 302 (less than 50 percent) in the experimental
> group. The diet of the other subjects who completed the
> study is unknown. This raises questions about the role of
> diet in explaining the results reported for recurrent
> coronary events.

This is really only a problem because of Hawthorne effect. Had
the two groups received equally strong dietary conunselling,
then even without dietary data at the end of the study, we
could conclude that differences between the two groups were
due differences in their diets rather than differential
behavioral changes brought about by unequal treatment.

> They seem to be saying that the problems in both the design
> of the study and the data gathering render its conclusions
> suspect. If that is the case, what other well-designed
> studies have been done to overcome those difficulties and
> make the diet as universally recommendable as it is today?

To my knowledge none have been done, and they need to be. In
particular, studies are needed that directly compare the
Mediterranean diet with the step 2 diet using hard disease
endpoints. We are in the middle of a paradigm shift, with
low-fat diets gradually being replaced by diets resembling
Mediterranean diets. This shit is largely based on studies of
effects of diet on intermediate endpoints such as blood
lipids, without the benefit of studies on actual heart
disease risk.

-Jay

[Paul Chefu]

On Tue, 13 Aug 2002 18:15:03 -0700, Jay Tanzman
<jtanzman@sph.llu.edu> wrote:

>
>
>Paul Chefurka wrote:
>>
>> The "Mediterranean" diet gets a lot of good notice for
>> being healthy overall, and especially good for the
>> prevention of heart disease. The Lyon Diet Heart Study is
>> usually referenced as the definitive finding on this type
>> of diet.
>
>It is not usually cited as definitive.

Maybe "definitive" is the wrong word. I've certainly hear the
study used as primary support for the efficacy of the
Mediterranean diet, though in retrospect that was more by
popularizers than scientists.

Thanks for the analysis, Jay. It looks as though we're in the
midst of a period of turmoil on the dietary front, as more and
better studies are done on a variety of approaches. That's
good to see.

Paul

[John 'The]

Once upon a time, our fellow Paul Chefurka rambled on about
"Re: Lyon Diet Heart Study questions." Our champion
De-Medicalizing in sci.med.nutrition retorts, thusly ...

>Maybe "definitive" is the wrong word. I've certainly hear the
>study used as primary support for the efficacy of the
>Mediterranean diet, though in retrospect that was more by
>popularizers than scientists.

Up until a few months ago, the "Lyon Diet Heart Study" was
Jay's definitive study for promoting canola oil. Then one day
Jay actually decided to read it. It was only then, that the
"Lyon Diet Heart Study" fell out of favor with Jay. :-(
--
John Gohde, Achieving good Health is an Art, NOT a Science!
http://NaturalHealthPerspective.com/ The ONLY Frauds in Health
are those who couldn't care less about prevention. Beware of
anybody who brags about eating a lousy diet, eating
crispbread, being overweight, or about smoking!

[Paul Chefu]

On Wed, 14 Aug 2002 15:44:02 GMT, John 'the Man'
<DeMan[80]@hotmail.com> wrote:

>Up until a few months ago, the "Lyon Diet Heart Study" was
>Jay's definitive study for promoting canola oil. Then one day
>Jay actually decided to read it. It was only then, that the
>"Lyon Diet Heart Study" fell out of favor with Jay. :-(

Bullshit. I went to Google Groups, and found a post from Jay
back in January in which he expressed strong doubts about the
Lyon methodology due to the Hawthorne effect. In fact, I read
everything that popped up on a search for "Tanzman Lyon" and
couldn't find a single post that supports your contention. I
did find a couple of threads in which he spanked your butt
over trans-fats, which might go some ways toward explaining
why you're so eager to slander him.

Or you might just be a fruitcake.

Paul

[John 'The]

Once upon a time, our fellow Paul Chefurka rambled on about
"Re: Lyon Diet Heart Study questions." Our champion
De-Medicalizing in sci.med.nutrition retorts, thusly ...

> I went to Google Groups, and found a post from Jay back in
> January in which he expressed strong doubts about the Lyon
> methodology due

Bingo!
--
John Gohde, Achieving good Health is an Art, NOT a Science!
http://NaturalHealthPerspective.com/ The ONLY Frauds in Health
are those who couldn't care less about prevention. Beware of
anybody who brags about eating a lousy diet, eating
crispbread, being overweight, or about smoking!

[John 'The]

Once upon a time, our fellow Paul Chefurka rambled on about
"Re: Lyon Diet Heart Study questions." Our champion
De-Medicalizing in sci.med.nutrition retorts, thusly ...

>On Wed, 14 Aug 2002 15:44:02 GMT, John 'the Man'
><DeMan[80]@hotmail.com> wrote:

>>Up until a few months ago, the "Lyon Diet Heart Study" was
>>Jay's definitive study for promoting canola oil. Then one
>>day Jay actually decided to read it. It was only then, that
>>the "Lyon Diet Heart Study" fell out of favor with Jay. :-(

>Bullshit.

"the Lyon Heart Study, the most effective diet study on the
prevention of heart disease" http://groups.google.com/groups?-
q=Tanzman+Lyon&hl=en&lr=lang_en&ie=UTF-8&oe=UTF-8&newwindow=1-
&safe=off&selm=3C448374.15705737%40sph.llu.edu&rnum=4

Okay drop "promoting canola oil" and substitute "an artificial
margarine was the main intervention," aka canola margarine.

And, I repeat, Jay Tanzman stated: "the Lyon Heart Study, the
most effective diet study on the prevention of heart disease."
Means the same thing as "the most definitive study" to me. :-)
--
John Gohde, Achieving good Health is an Art, NOT a Science!
http://NaturalHealthPerspective.com/ The ONLY Frauds in Health
are those who couldn't care less about prevention. Beware of
anybody who brags about eating a lousy diet, eating
crispbread, being overweight, or about smoking!