Re: Fitness versus fatness, round ?

[Guest guest]

In a message dated 10/15/04 8:16:04 PM Eastern Daylight Time, apater@... writes:

But look at the 95% confidence interval on the

data. The spread is quite large, and in the case of the fit men,

spans the 1.00 hazard ratio (0.55 - 1.37). I believe this can be

interpreted to mean that we can only say with 95% confidence that the

real risk of death for fit-fat Russian men is somewhere in the range

of 55% to 137% of fit-slim Russian men. Clearly this isn't a very

strong statement.

Al, when I saw your first post, I took that as one of the many odd things about that study. If you've got the stat knowledge, please elaborate. What you say about the confidence interval is exactly right, from what I know recently. Studies from years past used a different approach.

This study seems to say to me that the ratios given can't really be relied upon, for the Russian group anyway. If you can read full-text, was the sample size very small?

The conclusion that fat Rusians are better of than fit Russians is also odd. Differences in Russians? Just from seeing the news, the big two would be very high stress and also high alcohol.

--

Ken

[Guest guest]

In a message dated 10/15/04 3:10:09 PM Eastern Daylight Time, apater@... writes:

See the latest that I have seen on the fitness versus fatness

story.

What is going on in the Russians that may differ?

It being a prospective study comforted.

also, maybe it's just a poorly designed and executed study. Is this journal held in high esteem? I've heard the view that not all journals adhere to high standards, and so one must consider the source.

--

Ken

[Guest guest]

Hi All,

The paper appears to contain several interesting

findings:

1) In American men, being thin seemed to lead to a longer life,

independent of fitness.

2) In Russian men, being fat seemed to lead to longer life,

regardless of whether one was fit or not.

3) Being fit increased the lifespan of American and Russian men,

independent of weight.

Two observations are related to these results:

1) Finding #2 (fat is better than thin for Russian longevity)

obviously seems weird. But look at the 95% confidence interval on the

data. The spread is quite large, and in the case of the fit men,

spans the 1.00 hazard ratio (0.55 - 1.37). I believe this can be

interpreted to mean that we can only say with 95% confidence that the

real risk of death for fit-fat Russian men is somewhere in the range

of 55% to 137% of fit-slim Russian men. Clearly this isn't a very

strong statement. It suggest to me the fit-fat advantage in these men

may not be statistically robust, and therefore may not mean much if

anything.

2) Previous studies discussed have suggested being fit and fat is

better than being unfit and slim from a longevity perspective. This

study seems to suggest that the two conditions are pretty much

equivalent, with fit and slim being quite a bit better still (~40%

better). Note the confidence intervals for the US men are quite a bit

smaller than those for the Russians, suggesting greater significance.

Cheers, Al Pater.

--- In , " old542000 " <apater@m...>

wrote:

>

> Hi All,

>

> See the latest that I have seen on the fitness versus fatness

> story.

>

> What is going on in the Russians that may differ?

>

> It being a prospective study comforted.

>

> Int J Obes Relat Metab Disord. 2004 Sep 14 [Epub ahead of print]

> Associations of fitness and fatness with mortality in Russian and

> American men

> in the lipids research clinics study.

> s J, Evenson KR, O, Cai J, R.

>

> ...Prospective closed cohort. SUBJECTS:: 1359 Russian men and 1716

US

> men aged

> 40-59 y at baseline (1972-1977) ... body mass index (BMI) ...

> proportional

> hazard models that

> included covariates for age, education, smoking, alcohol intake and

> dietary keys

> score. All-cause and cardiovascular disease (CVD) mortality were

> assessed

> through 1995.

> ...In Russian men, fitness was associated with all-cause

> and CVD mortality, but fatness was not. For mortality from all

> causes, compared

> to the fit-not fat, the adjusted hazard ratios were 0.87 (95% CI:

> 0.55, 1.37)

> among the fit-fat, 1.86 (95% CI: 1.31, 2.62) among the unfit-not

fat

> and 1.68

> (95% CI: 1.06, 2.68) among the unfit-fat. Among US men, the same

> hazard ratios

> were 1.40 (95% CI: 1.07, 1.83), 1.41 (95% CI: 1.12, 1.77) and 1.54

> (95% CI:

> 1.24, 2.06), respectively. There were no statistically significant

> interactions

> between fitness and fatness in either group of men for all-cause or

> CVD

> mortality.

> ... The effects of fitness on mortality may be more robust

> across populations than are the effects of fatness.

>

> PMID: 15365584 [PubMed - as supplied by publisher]

>

> Cheers, Al Pater.