OBESITY LINKED TO HIGHER RATES OF CHRONIC ILLNESS

Thank you for this important information. With your permission I've
cross posted to alt.support.diet.low-carb and
alt.support.diabetes/misc.health.diabetes



Oink, Oink, Oink!!! wrote:
Contact: Jess Cook
Phone: 310-451-6913
Fax: 310-451-6988
Email:

RAND
1700 Main Street
P. O. Box 2138
Santa Monica, CA 90407-2138
http://www.rand.org

1200 South Hayes Street
Arlington, Virginia 22202-5050
703-413-1100

OBESITY LINKED TO HIGHER RATES OF CHRONIC ILLNESS AND WORSE PHYSICAL
QUALITY OF LIFE THAN SMOKING, DRINKING OR POVERTY
THREE OF FIVE ADULT AMERICANS ARE OVERWEIGHT OR OBESE
SANTA MONICA, Calif., June 7 - How serious a health problem is obesity?
In a report published in the current issue of the British journal
Public Health, two RAND researchers find that obesity is associated
with higher rates of chronic medical conditions and with worse physical
health-related quality of life than are lifetime smoking, problem
drinking or poverty.

Their study, the first to compare the effects of these major health
risks on morbidity, also provides new estimates indicating that three
of every five adult Americans are either overweight (36 percent) or
obese (23 percent). In fact, far more people now are overweight or
obese than are, collectively, daily smokers, problem drinkers and below
the federal poverty line.

The study is likely to reinforce the growing concern that obesity has
become epidemic. It also underscores the top priority the Surgeon
General is giving this year to developing a national action plan to
reduce obesity's prevalence.

"Americans haven't given overweight the same attention as other risks,
like smoking, but it is clearly a top health problem and one that is on
the rise in all segments of the population. More effective clinical and
public health approaches are urgently needed," declare co-authors
Roland Sturm, a RAND economist, and Kenneth Wells, a psychiatrist on
the RAND staff who is also on the faculty of the UCLA School of
Medicine.

Sturm and Wells analyzed data from a nationally representative,
household telephone survey of 9,585 adults conducted in 1998. The
survey included questions covering height, weight, income, smoking and
drinking habits and health status.

The researchers used the body mass index (BMI), a ratio of height to
weight, to define overweight and obesity. (For example, someone who is
5 ft. 4 in. tall and weighs 174 lbs. or more or 5 ft. 10 in. and 209
lbs. or more is considered obese). They measured health status by
analyzing responses to questions concerning some 17 major chronic
conditions and 12 quality-of-life issues.

The study was funded by the Robert Wood Johnson Foundation. RAND is a
nonprofit organization that improves policy and decisionmaking through
research and analysis.


##
For a copy of the journal article, contact RAND customer service at
.

In article .com,
Sweet Zombie Jesus! wrote:
Thank you for this important information. With your permission I've
cross posted to alt.support.diet.low-carb and
alt.support.diabetes/misc.health.diabetes

There is PROBABLY a causal effect, but this cannot be
deduced from the data presented, or any similar data.

Determining causal relations from non-controlled data
may eventually be possible, but not from looking at
straightforward associations. The damage from fire
is associated with the number of fire trucks brought
to the scene, but does this mean there would be less
damage if fewer trucks were brought in?
--
This address is for information only. I do not claim that these views
are those of the Statistics Department or of Purdue University.
Herman Rubin, Department of Statistics, Purdue University
Phone: (765)494-6054 FAX: (765)494-0558

I'm not sure I would agree with this assessment. The study is based on
rank of correlation and covariance not via simple regression. (I may be
wrong; most of the stats I use in my work are of the 'simple/inear'
variety)

/Herman Rubin wrote:
/ In article .com,
/ Sweet Zombie Jesus! wrote:
/ Thank you for this important information. With your permission I've
/ cross posted to alt.support.diet.low-carb and
/ alt.support.diabetes/misc.health.diabetes
/
/ There is PROBABLY a causal effect, but this cannot be
/ deduced from the data presented, or any similar data.

/ Determining causal relations from non-controlled data
/ may eventually be possible, but not from looking at
/ straightforward associations. The damage from fire
is associated with the number of fire trucks brought
to the scene, but does this mean there would be less
damage if fewer trucks were brought in?
--
This address is for information only. I do not claim that these views
are those of the Statistics Department or of Purdue University.
Herman Rubin, Department of Statistics, Purdue University
Phone: (765)494-6054 FAX: (765)494-0558

In article . com,
Sweet Zombie Jesus! wrote:
I'm not sure I would agree with this assessment. The study is based on
rank of correlation and covariance not via simple regression. (I may be
wrong; most of the stats I use in my work are of the 'simple/inear'
variety)

This is just as much a misuse of statistics. Correlation
is NOT causation; it can just as well go completely the
other way. In fact, correlation is worse than regression,
as other variables cannot be considered. Partial correlation
is essentially regression.

That is why controlled experiments are used. But on this
matter, we cannot use controlled experiments; we cannot
tell some people to eat more, and others less. Also, some
people become obese on diets on which others remain thin.
There is evidence that the RDAs for different people should
vary by a factor of up to 10.

Possibly we could do more if we had an excellent quantitative
theory of biological activity, including the genetic component.
But we are anywhere even near this.

/Herman Rubin wrote:
/ In article .com,
/ Sweet Zombie Jesus! wrote:
/ Thank you for this important information. With your permission I've
/ cross posted to alt.support.diet.low-carb and
/ alt.support.diabetes/misc.health.diabetes
/
/ There is PROBABLY a causal effect, but this cannot be
/ deduced from the data presented, or any similar data.

/ Determining causal relations from non-controlled data
/ may eventually be possible, but not from looking at
/ straightforward associations. The damage from fire
is associated with the number of fire trucks brought
to the scene, but does this mean there would be less
damage if fewer trucks were brought in?
--
This address is for information only. I do not claim that these views
are those of the Statistics Department or of Purdue University.
Herman Rubin, Department of Statistics, Purdue University
Phone: (765)494-6054 FAX: (765)494-0558

Herman Rubin wrote:
In article . com,
Sweet Zombie Jesus! wrote:
I'm not sure I would agree with this assessment. The study is based
on rank of correlation and covariance not via simple regression. (I
may be wrong; most of the stats I use in my work are of the
'simple/inear' variety)

This is just as much a misuse of statistics. Correlation
is NOT causation; it can just as well go completely the
other way. In fact, correlation is worse than regression,
as other variables cannot be considered. Partial correlation
is essentially regression.

That is why controlled experiments are used. But on this
matter, we cannot use controlled experiments; we cannot
tell some people to eat more, and others less. Also, some
people become obese on diets on which others remain thin.
There is evidence that the RDAs for different people should
vary by a factor of up to 10.

Well, we could do some animal studies. Has anyone seen good animal studies
for Type 2 and correlations of body mass, diet, and Type 2 in animals?


Possibly we could do more if we had an excellent quantitative
theory of biological activity, including the genetic component.
But we are anywhere even near this.

/Herman Rubin wrote:
/ In article
.com, / Sweet
Zombie Jesus! wrote: / Thank you for
this important information. With your permission I've / cross
posted to alt.support.diet.low-carb and /
alt.support.diabetes/misc.health.diabetes /
/ There is PROBABLY a causal effect, but this cannot be
/ deduced from the data presented, or any similar data.

/ Determining causal relations from non-controlled data
/ may eventually be possible, but not from looking at
/ straightforward associations. The damage from fire
is associated with the number of fire trucks brought
to the scene, but does this mean there would be less
damage if fewer trucks were brought in?

In article ,
Nico Kadel-Garcia wrote:
Herman Rubin wrote:
In article . com,
Sweet Zombie Jesus! wrote:
I'm not sure I would agree with this assessment. The study is based
on rank of correlation and covariance not via simple regression. (I
may be wrong; most of the stats I use in my work are of the
'simple/inear' variety)

This is just as much a misuse of statistics. Correlation
is NOT causation; it can just as well go completely the
other way. In fact, correlation is worse than regression,
as other variables cannot be considered. Partial correlation
is essentially regression.

That is why controlled experiments are used. But on this
matter, we cannot use controlled experiments; we cannot
tell some people to eat more, and others less. Also, some
people become obese on diets on which others remain thin.
There is evidence that the RDAs for different people should
vary by a factor of up to 10.

Well, we could do some animal studies. Has anyone seen good animal studies
for Type 2 and correlations of body mass, diet, and Type 2 in animals?

You can, but while many of these animal studies yield
important information, they are more accurate because
they typically use inbred strains, but this keeps them
from getting the appropriate data depending on individual
variation. Also, diabetes and metabolic syndrome diseases
in animals are chemically induced, and not due to the
genetic adaptations made by man. Your mileage may vary
is the "slogan" of this group, and using an inbred strain
of mice does not achieve this.

Furthermore, there are mental effects, largely ignored by
medical people. These are usually not identifiable in
laboratory animals.


Possibly we could do more if we had an excellent quantitative
theory of biological activity, including the genetic component.
But we are anywhere even near this.

/Herman Rubin wrote:
/ In article
.com, / Sweet
Zombie Jesus! wrote: / Thank you for
this important information. With your permission I've / cross
posted to alt.support.diet.low-carb and /
alt.support.diabetes/misc.health.diabetes /
/ There is PROBABLY a causal effect, but this cannot be
/ deduced from the data presented, or any similar data.

/ Determining causal relations from non-controlled data
/ may eventually be possible, but not from looking at
/ straightforward associations. The damage from fire
is associated with the number of fire trucks brought
to the scene, but does this mean there would be less
damage if fewer trucks were brought in?







--
This address is for information only. I do not claim that these views
are those of the Statistics Department or of Purdue University.
Herman Rubin, Department of Statistics, Purdue University
Phone: (765)494-6054 FAX: (765)494-0558

Herman Rubin wrote:
In article ,
Nico Kadel-Garcia wrote:
Herman Rubin wrote:
In article . com,
Sweet Zombie Jesus! wrote:
I'm not sure I would agree with this assessment. The study is based
on rank of correlation and covariance not via simple regression. (I
may be wrong; most of the stats I use in my work are of the
'simple/inear' variety)

This is just as much a misuse of statistics. Correlation
is NOT causation; it can just as well go completely the
other way. In fact, correlation is worse than regression,
as other variables cannot be considered. Partial correlation
is essentially regression.

That is why controlled experiments are used. But on this
matter, we cannot use controlled experiments; we cannot
tell some people to eat more, and others less. Also, some
people become obese on diets on which others remain thin.
There is evidence that the RDAs for different people should
vary by a factor of up to 10.

Well, we could do some animal studies. Has anyone seen good animal studies
for Type 2 and correlations of body mass, diet, and Type 2 in animals?


I agree with your second point vis a vis 'mental effects'. However we
do not typically use inbred strains for metabolic studies (mainly due
to the expense e.g, an inbred AJ57 mouse costs 6-7 times as much as a
outbred CD1).

/ You can, but while many of these animal studies yield
/ important information, they are more accurate because
/ they typically use inbred strains, but this keeps them
/ from getting the appropriate data depending on individual
/ variation. Also, diabetes and metabolic syndrome diseases
/ in animals are chemically induced, and not due to the
/ genetic adaptations made by man. Your mileage may vary
/ is the "slogan" of this group, and using an inbred strain
/ of mice does not achieve this.

/ Furthermore, there are mental effects, largely ignored by
/ medical people. These are usually not identifiable in
/ laboratory animals.


Possibly we could do more if we had an excellent quantitative
theory of biological activity, including the genetic component.
But we are anywhere even near this.

/Herman Rubin wrote:
/ In article
.com, / Sweet
Zombie Jesus! wrote: / Thank you for
this important information. With your permission I've / cross
posted to alt.support.diet.low-carb and /
alt.support.diabetes/misc.health.diabetes /
/ There is PROBABLY a causal effect, but this cannot be
/ deduced from the data presented, or any similar data.

/ Determining causal relations from non-controlled data
/ may eventually be possible, but not from looking at
/ straightforward associations. The damage from fire
is associated with the number of fire trucks brought
to the scene, but does this mean there would be less
damage if fewer trucks were brought in?







--
This address is for information only. I do not claim that these views
are those of the Statistics Department or of Purdue University.
Herman Rubin, Department of Statistics, Purdue University
Phone: (765)494-6054 FAX: (765)494-0558

On 3 Aug 2006 08:49:57 -0700, "Sweet Zombie Jesus!"
Huffed and Puffed the following into the
madness of usenet:

thank zombie's ISP:
OrgName: GTIS
OrgID: GTIS
Address: PWGSC/GTIS PDP III 3A2
City: Hull-Gatineau
StateProv: QC
PostalCode: J8X-4A6
Country: CA

NetRange: 198.103.0.0 - 198.103.255.255
CIDR: 198.103.0.0/16
NetName: NETBLK-CDAGOVN-C
NetHandle: NET-198-103-0-0-1
Parent: NET-198-0-0-0-0
NetType: Direct Allocation
NameServer: RELAY.SRV.GC.CA
NameServer: RUSTY.SRV.GC.CA
Comment:
RegDate: 1993-03-25
Updated: 2003-05-05

RTechHandle: HOSTM227-ARIN
RTechName: Hostmaster
RTechPhone: +1-819-956-1180
RTechEmail:

OrgTechHandle: HOSTM227-ARIN
OrgTechName: Hostmaster
OrgTechPhone: +1-819-956-1180
OrgTechEmail:

So....."Ma¢k" what's does posting my ISP have to do with our lab using
outbred vs inbred strains of mice for metabolic studies or my use of
linear vs non-parametric stats? Why don't you read the posted studies,
analyse the data and make a positive contribution to the discussion for
a change. You're such a pathetic little idiot, grow up!

?Ma¢k wrote:
? On 3 Aug 2006 08:49:57 -0700, "Sweet Zombie Jesus!"
? Huffed and Puffed the following into
the
? madness of usenet:
?
? thank zombie's ISP:
OrgName: GTIS
OrgID: GTIS
Address: PWGSC/GTIS PDP III 3A2
City: Hull-Gatineau
StateProv: QC
PostalCode: J8X-4A6
Country: CA

NetRange: 198.103.0.0 - 198.103.255.255
CIDR: 198.103.0.0/16
NetName: NETBLK-CDAGOVN-C
NetHandle: NET-198-103-0-0-1
Parent: NET-198-0-0-0-0
NetType: Direct Allocation
NameServer: RELAY.SRV.GC.CA
NameServer: RUSTY.SRV.GC.CA
Comment:
RegDate: 1993-03-25
Updated: 2003-05-05

RTechHandle: HOSTM227-ARIN
RTechName: Hostmaster
RTechPhone: +1-819-956-1180
RTechEmail:

OrgTechHandle: HOSTM227-ARIN
OrgTechName: Hostmaster
OrgTechPhone: +1-819-956-1180
OrgTechEmail: